Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: Johnansaro on March 02, 2018, 12:16:11 PM
-
Hi there
what is exactly the meaning of electronic gear ratio in Delta asda b2 Servo Driver?
and how to calculate it?
regards
jhonansaro
-
Hi,
if your servo encoder is 10,000 count per rev then Mach would have to produce 10,000 pulses in one second to cause the servo to turn one revolution.
If you want the servo to run at 3000 rpm it will have to run at 50 revolutions per second and Mach would have to produce 500,000 pulses per second to
do it. A parallel port installation can't even get close to 500,000 pulses per second, 25,000 per second is the norm.
Thus because of the limited pulse speed of Mach you can't use the full potential of the servo. One way around that is to use electronic gearing.
If the denominator is set to 20 then for each pulse received by the drive from Mach the servo will turn 20 counts of its encoder. If Mach produces its
max 25,000 pulses per second the servo will turn 20 *25,000=500,000 counts per second ie 3000 rpm.
You don't get it for nothing. Without electronic gearing the servo resolution is 1/10,000 rev=0.036 degrees or 2.16 min of arc. With an electronic gear ratio of 1/20
the resolution is 20/10,000 rev=0.72 degree or 43.2 min of arc. Using electronic gearing trades resolution for speed.
Most external motion controllers have pulse speeds much higher than Machs parallel port, 100kHz is a minimum, my ESS can manage 4MHz. Most servo drives
can't even with differential signaling accept more than 500kHz so 4MHz is somewhat academic. It is still useful if reading an encoder however. Many Delta servos
for instance have 17 bit encoders for 131,072 counts per rev or 6.5MHz at 3000 rpm!
Give me some details about what your trying to achieve and I'll try to help with the two numbers you require (numerator and denominator). You'll need to know the maximum
useful signaling rate of your controller and the maximum speed that you wish to run the servo. Note that single ended signaling is good to about 75kHz (the manual
might say more, but 75Khz is safely conservative) and beyond that you'll need differential signaling. If your BoB can't do it you'll have to make a circuit with a line
driver IC on it.
Craig
-
Hi,
just realized that I don't know what servo you are using, or in particular what is the encoder count per rev?
Craig
-
Hi joeaverage
Many thanks for the reply, i have a 750 w delta servo motor, and delta adas b2 driver. as you said my driver has a 17 bit encoders for 160,000 counts per rev.
And my BOB is a Mach3 Motion Card (ESS- Ethernet Smooth Stepper----HM-LAN-4Axis)
The physical Top Speed that I require is a bout 12,000 mm/min or 3200 Rev/min.
I have done some math, Which off course I am not so sure of, and they are as Follow:
(No. tooth)=25 ----- (Pitch Dist. mm) =7 linear Travel 175 mm
Encoder 160000
P1-44 5000 (Neominator)
P1-45 10 (denominator)
From the above If I have a 10:1 Gear Box Then with 3200 rev/min I will get a actual 320 rev/min and every revolution at 175 mm I will end up with 56,000 mm/mi
Therefore if I increase the value of p1-44 I can Lower the speed. by the way with above assumption the value for Steps per Unit in mach3 should be
about 4.57
Could you verify all that
regards
jhonansaro
-
Hi,
still trying to digest the details but that encoder count doesn't sound right.
A 17 bit encoder I think has a count of 131,072 per rev. Double and triple check the manual
about that count number.
Craig
-
Hi,
if I understand your numbers the axis moves 175mm per revolution of the gear.
If you want 12000mm/min travel then:
12000/175=68.7514 rev per min of the gear, or 685.714 revs per min of the input shaft of your 10:1 reducer,
ie 658.714 rpm of your servo.
At 131072 pulse per rev:
658.714 *131072 /60=1,497,965 pulse per second. Thats too fast to signal your servo, I would recommend reducing
the pulse rate required using electronic gearing by a factor of ten:
1,497,965 / 10=149,796 pulse per second. Thats about 150kHz. You could, at a stretch, do that with single
ended signalling.
This would require a denomintator of 10 and a numerator of 1. Easy!
-
Hi,
with the numbers I've proposed the 'step per mm':
1mm / 175mm gear revs=0.00571428 gear revolutions/mm
0.00571428 * 10=0.0571428 reducer input shaft revs/mm =0.0571428 servo revs/mm
0.0571428 *131,072=7489.82 encoder counts per mm
7489.82 / 10 (electronic gearing denom.)=748.98 Mach pulses/mm
Rounding 'steps per mm'=749 or a resolution of 1.33um, pretty damn fine!
Craig
-
Hi,
have been reading the manual and it turns out you can program the encoder count. Page 7-40.
P1-46▲ GR3 Encoder Output Pulse Number Address: 015CH
015DH
Operation
Interface: Keypad/Software Communication Related Section: N/A
Default: 2500
Control
Mode: ALL
Unit: pulse
Range: 4 ~ 40000
Data Size: 32-bit
Display
Format: Decimal
Settings:
This parameter is used to set the pulse numbers of encoder outputs per motor
revolution.
Please note: When the following conditions occur, the output frequency for pulse
output may exceed the specification and cause that the servo drive fault AL018
(Encoder Output Error) is activated
Note the default is 2500 ppr which corresponds to an encoder count of 10,000 counts per rev.
My suggestion is stick with that.
The calculations change:
12000/175=68.571428
68.571428 *10=685.7142
685.7142 *10,000/60=114285.7142 pulse per second. or 114.2855kHz
ie You don't need electronic gear reduction at all.
The 'steps per mm':
1 /175=0.005714285
0.005714285 *10=0.05714285
0.05714285 *10000=571.428 pulse per mm
This is easier again. Program the servo to produce 10000 count (2500 ppr), which is its default programming. As you only require 685 rpm to reach
your target speed with a 10:1 reducer you don't need electronic gearing. The maximum pulse rate required is 114 kHz and within the drives singled ended
signaling capacity, easy.
Craig
-
Hi Craig
Excellent job, your detailed explanation is awesome. thanks again.
Ok from your calculation is clear that if I want to have a speed of 12000 mm/min then the motor must have around 685 rev/min, but do not you agree
That I will get better performance and more torque at speed around 3000 rev/min, therefore I should have a 4:1 electronic gear ration in order to force the motor
to run at higher speed of 4*685=2740 ?
also again I am quoting from your calculations that with (685.7142 *10,000/60=114285.7142 pulse per second. or 114.2855kHz) ,
Then how about the mach3 kernel speed which is around 45 kHz, isn't this some what over mach3's kernel speed?
regards
jhonansaro
KHz
-
Hi,
if your machine can handle speeds approaching 1m per second and handle the G forces involved then
you have a pro machine and should not be pissing around with Mach.
Who cares abouts Machs kernel? You are running an ESS, its 'kernel' if thats what you want to call it is 4Mhz.
Without electronic gearing you could still make the axis go four times faster just by increasing the pulses per second
from 114k to 456k. Differential signalling required but no change in electronic gearing required.
The torque from an AC servo is indepenednt of speed, it has no more torque at high speed than at low speed. So I don't
agree you'll get any more performance by going faster. In fact unless you plan the decceleration phase properly
and have the servo fault on decel then you risk a crash.
-
Hi,
should explain..Machs 'kernel' refers to the parallel port pulse engine. As you are using an ESS you are not using
Machs parallel port and therefore its pulse speed limitation is irrelevant.
therefore I should have a 4:1 electronic gear ration in order to force the motor
to run at higher speed of 4*685=2740 ?
If you run the motor at four times the speed the axis will move four times as fast UNLESS you use a different reduction
gearbox. "Electronic gearing' has no bearing on that.
Craig
-
Hi Craig
I would like to thank you for your detailed replies, and explanations
I should mention that servo motors are new to me, and there are a lot that I need to learn.
All along i was imagining that one can control the speed (rpm) by modifying Electronic Gearing, apparently that is not the case, I tried to find some
Reading on this issue, but it is very hard to find good reading on it. if you have some good reading or clips, please let me know.
Apparently with my Servo motor, it's driver and pinion setup, I should make it easy for my self, and i do not need to Change or modify much.
I have movements, and Electronic gear is Ok as it is, off course I should mention that as you said the default value for
P1-44 16 Numerator
P1-45 10 denominator
Do I need to change this value to both (10, or 16) or should I leave it as it is?
Also is there any thing else that I need to change in order to get better performance out of my motor/
regards
jhonansaro
-
Hi,
by my calculation you don't need any electronic gearing at all.
Denominator=Numerator =1
Would suggest you enter max speed of 800 rpm say. Keep the max acceleration somewhat less than the absolute theoretical best. The real challenge with
a servo comes when you are decelerating. The motor turns into a generator and energy comes from the servo back into the servo drive and they don't
like it. There may be optional braking resistors for your servo to improve the situation. If you try to decelerate too quickly the drive will fault 'bus overvoltage'
and the servo will coast to a stop. If it coasts right off the end of the axis....to bad....should've been more careful planning the deccel phase.
Modern AC servos are extremely capable, if you can only get 10% of best out of one it will still be amazing. Don't worry about it, get it set it up and
use/experiment with it, you'll be really impressed. The trick is to be a little conservative about accel/deccel, they are so capable that you are inclined to keep
upping the ante only to find that you break your machine.
Craig
-
Hi Craig
On the issue of electronic gearing, today i was experimenting with my Servo, and with low value of P1-44 (Numerator Gear Ratio) set like 10 or 16, with value of P1-45(Denominator) set
at value of 10. I will get barely any movement, But as I increase the P1-44 to some thing like 160, I will get a good RPM out of the servo.
In conclusion it appears that by increasing the Numerator I can increase the RPM, but you said that they are not related. How could you explain this?
I should mention that I was using mach3 Motor tuning in order to increase the speed. but with low values of P1-44 no matter how much i increase the speed in mach3 I would not gain
Much RPM, I get some but very little. Basically with speed set to some value like 1000 rpm/min in mach3 has different meaning with Low P1-44 vs High values for p1-44
regards
jhonansaro
-
Hi,
electronic gearing is a means of controlling or setting the number of input pulses necessary to produce a
movement of a number of encoder counts.
If you had a signal generator producing a steady 1000 pules per second then with no gearing (default
10,000 count encoder) it would move 1/10 rev per second.
As you increase the electronic gearing it will get faster, that is that the same 1000 pulses will cause an
encoder count movement of 20,000 say or 2 revs per second.
You can set both nominator and denominator equal and that means that one input pulse causes movement
of one encoder count. Whether you set them both at 16 (or 10 or 1 or 57....) doesn't matter.
If you are going to do some testing use Mach3 to produce the pulses but don't alter it otherwise you can't be
sure what it is thats changing.
Craig
-
Hi,
try this from the manual:
P1-45▲ GR2 Electronic Gear Ratio (Denominator) (M) Address: 015AH
015BH
Operation
Interface: Keypad/Software Communication Related Section:
Section 6.2.3
Default: 10
Control
Mode: PT
Unit: pulse
Range: 1 ~ (231-1)
Data Size: 32-bit
Display
Format: Decimal
Settings:
This parameter is used to set the denominator of the electronic gear ratio. The
numerator of the electronic gear ratio is set by P1-44. P2-60 ~ P2-62 are used to set
the additional numerators.
As the wrong setting may cause motor to run chaotically (out of control) and it may
lead to personnel injury, therefore, ensure to observe the following rule when setting
P1-44, P1-45.
The electronic gear ratio setting (Please also see P1-44, P2-60 ~ P2-62):
f1 f2 = f1 x N
M
N
M
Pulse input
Position
command
The electronic gear ratio setting range must be within: 1/50<N/M<25600.
Please note:
1. In PT mode, the setting value of P1-45 can not be changed when the servo drive
is enabled (Servo On).
The electronic gear function provides easy travel distance ratio change. However, the over
high electronic gear ratio will command the motor to move not smoothly. At this time, the users
can use low-pass filter parameter to improve this kind of situation. For example, assume that
the electronic gear ratio is equal to 1 and the encoder pulse per revolution is 10000ppr, if the
electronic gear ratio is changed to 0.5, then the motor will rotate one pulse when the command
from external controller is two pulses.
For example, after the proper electronic gear ratio is set, the reference travel distance is 1
Note that the formula expressed is not displayed correctly in the quote. Read the manual.
Craig
-
Hi Craig
Well your last reply was just over my head, and with my current state of concentration I do not think I can digest it any time soon.
But my setting currently for my servo are as follow, and I should say that I am getting a smooth run so far, but off course i have not install them completely yet;
P1-44 480
P1-45 16
P1-36 5 ms
P1-52 100
P2-08 36
Could you please verify all these for me
Finally pulse per unit in mach3 is set at 571 , "as you calculated before "
regards
jhonansaro
-
Hi,
did you read the page 6-8 in the manual?
It doesn't appear so.
Currently you have the numerator =480 and denominator at 16.
One input pulse will result in 480/16=30 encoder counts, not what you want, you want one input pulse to effect one encoder count, ie no electronic gearing
at all.
Make P1-44 equal to P1-45, I don't really care what, if you set one as 16 then set the other as the same.
The formula on page 6-8:
F2= F1 X (M/N)
You want F1=F2, therefore M/N must equal one, therefore M must equal N.
Craig
-
Hi Craig
thanks for the reply.
are you telling me that the Electronic gear Ratio(EGR) must be equal 1 in order to satisfy the formula? then what is the use of having such ratio?
Also I did what you asked me, and I set the value of
P1-44 =16
P1-45 =16
But even Thu in mach3 motor tuning i set the speed at 10000 mm/min, my speed has been reduced considerably to some thing like 1 rev/sec, which seems
not bad 1*60= 60 rev/min and 60*175=10500 mm But please be reminded that I have a 10:1 GearBox so I think I need to Increase my speed 10 times more?
When I had :
P1-44 =480
P1-45 =16
My speed was much higher, I do not have the means to calculate (Or may be I have through the use of driver and/or the software, But I just do not know how to)
But i could say that my speed was to some thing like over 10 rev/sec, which is what i need.
Could you verify all these to me?
on the issue of checking my actual speed, do you know how can find out about from the software of the driver, (you can get the software form the net, or I could send it to you)
regards
jhonansaro
-
Hi,
the point of having an electronic gear ratio is that if you have a controller that has a slow output pulse rate that it can still be used to provide the
fast encoder count required by the servo. Just because the manufacturer put electronic gearing there doesn't mean you have to use it.
With the electronic ratio set to 1 (numerator and denominator equal) if Mach/ESS produces 114,000 pulses per second your servo should spin 11.4 revs per
second or 684 rpm.
Note this assumes a 10000 count encoder. Make sure that this is the case.
Craig
-
Hi,
make sure P1-46 is 2500
Craig
-
Hi,
please post a screen shot of your motor tuning page.
Craig
-
hi Craig
thanks a million for your replies.
attached are prints of my Parameters.
I am sending them in 2 versions in case you may have any difficulty opening them up.
Also on the values of
P1-44
P1-45
I set them as your suggestions to the same number, and currently is set at 1600, which is only an experiment. By the way for these values,
What difference are between setting them a high value like 1600 vs. a low value such as 10 or 16?
I also noticed that the default value for them are 16, and 10
regards
jhonansaro
-
Hi Craig
also the attached text file is the list of parameters that Have been changed Vs default ones,
I thought it might be helpful
regards
jhonansaro
-
Hi,
those two parameters are numerator and denominator ie a fraction.
For instance 10/16 is exactly the same as 1000/1600 or 5/8. In this case you don't want any gearing so the fraction should be equal to 1 so:
1/1 or 10/10 or 16/16 or 64/64, I don't care which and, except going to extremes, neither does the drive.
Please make sure parameter P1-46 is 2500 and please post a screenshot of your motor tuning page.
Craig
-
Hi Craig
Have you looked at my parameters that i had sent you in my 2 previous replies?
What is your opinion about them?
as I said before with If i set the values of numerator and denominator to the same value like 16, and as you can see the value of P1-46 = 2500 already, but
The Motor shafts moves very slowly some thing like 1 rev/sec,
I think the problem is some where among the other parameters, do you have any recommendation that which parameters do i need to change
in order to get better or higher RPM?
regards
jhonansaro
-
Hi,
no I see nothing obvious so far. Can you post a screen pic of your motor tuning page?
Craig
-
175mm/rev Motor
Motor 3200U/min -> max Speed 560000mm/min -> 9333mm/s
P1 -44 -> 1
P1 -45 -> 1
P1 -46 -> 17500
Motor Tuning
steps per -> 100
velocity -> what ever your machine can handle before flying away
accelereation -> what ever your 750W Motor can handle with this big mechanikal Resolution(will be imho nearly nothing)
i am sure one of the given Parameters (imho 175mm/rev) is wrong.
-
Hi Thomas,
OP has a 10:1 reduction gearbox for 17.5mm per motor rev.
Craig
-
Hi,
P1 -46 -> 17500
Wrong, this is the ppr NOT the encoder count. If you truly wish 17500 as encoder count the parameter
must be 17500/4=4375/
Craig
-
Hi,
even if you did have the encoder set to a count of 17500 then the pulse rate would be:
17500 X 3000 /60=875kHz, thats low RF frequency, good luck trying to signal your drive!
Craig
-
Hi there
Thanks for your replies
I tried the following:
P1 -44 = 1
P1 -45 = 1
P1 -46 = 17500 also P1-46 = 4375
But nothing seems to Increase my speed, with all the above parameter settings, The speed remains sort of unchanged, and low some thing like 1 rev/sec
So far it appears the only thing that significantly Increase the speed is when P1-44 is about over 10 time the value of P1-45
attached is a pic. of my Motor Tuning
Also TPS:
How and where did you get these values (Motor 3200U/min -> max Speed 560000mm/min )
Isn't the 560000 mm/min kind of high value?
did you get 3200 from 560000/175?
Regards
jhonansaro
-
Hi,
kool, the motor tuning looks good. Now double and triple check your speed rate override. I suspect
you have overide in action which is also affecting the rapids.
I would go back to P1-46 =2500
Now MDI:
G0 x500
x0
x500
x0
What speed is the motor doing approx?
Craig
-
Hi,
So far it appears the only thing that significantly Increase the speed is when P1-44 is about over 10 time the value of P1-45
This is because you can't get Mach/ESS to produce a 114 kHz pulse stream. Why not...is the real question.
Craig
-
Hi there
I am back at 2500, and I tried the MDI Moves. I do not know what could I find out from these moves. Basicly It appears that Speed is always the same some thing less than a 1 Rev/sec.
But as I increase the Steps per unit in Motor tuning Dialog to some thing like 1200 the speed Increases to about 2 rev/sec.
But it appears that there is a ceiling on this, and when I try to double the Steps per unit to value like 2400 then in the Motor Tuning dialog Box the speed drops
Automatically and no mater if I try i can not increase it much more.
S/U=600 speed 10000 mm/min I can Increase the speed to 20000 mm/min and actually the shaft rev/sec doubles, but apparently I can not increase the speed to more than 23000 mm/min
S/U =1200 at the same speed of 10000 mm/min the actual speed of the shaft of the motor Doubles to 2 rev/sec , But then In mach3 Motor Tuning I can not increase the speed to more than 1141 mm/min.
I should mention that presently I have not mounted my Gearbox to the motor, and motor is moving with out the gear box, but i am planing to use this 10:1 gearbox in actual run.
Finally what parameter sets the speed over ride in Delta Driver?
regards
jhonansaro
-
Hi,
Finally what parameter sets the speed over ride in Delta Driver?
Don't know just yet, it is of secondary importance at the moment. I believe what is happening is that Mach/ESS/BoB is struggling to signal your servo driver.
What BoB are you using? Can you post the manual for it?
What I'm thinking it that the BoB may have slow speed optoisolators as outputs and may well not be fast enough. Another possibility is that the BoB may have
a 3.3V or 5V output whereas your driver may be expecting 24V inputs. A third possibility is that the BoB may have active low outputs say whereas the driver inputs are
active high say.
We are going to need documentation and good documentation at that to diagnose this. I believe that all the settings in the drive are either correct or close enough
for the moment. We are now looking for an issue with Mach/ESS/BoB.
Do you have or have access to an oscilloscope?
Craig
-
Hi,
just noticed a small detail with the motor tuning, the steps per is 587 in the pic and they should be 571. Small detail, its not whats stopping it from working.
When we do get it going then you'll want it accurate.
Craig
-
Hi,
if you are using a pretty standard BoB then you'll use diagram C3-1 on page 3-30. Note your BoB will require standing 24V at open collector.
Really, REALLY need the info about your BoB.
Craig
-
Hi,
just re-read one of your early posts and I'm confused. Do you have a Warp9TD EthernetSmoothStepper or not? That HM4...whatever is not a BoB that I can see, it
looks like a motion controller.
Craig
-
Also TPS:
How and where did you get these values (Motor 3200U/min -> max Speed 560000mm/min )
Isn't the 560000 mm/min kind of high value?
did you get 3200 from 560000/175?
from a post on the first side
From the above If I have a 10:1 Gear Box Then with 3200 rev/min I will get a actual 320 rev/min and every revolution at 175 mm I will end up with 56,000 mm/mi
ok i missed the 10:1 gearbox,sorry.
and 1 missed 4x on encoder
so we are here
17.5mm/rev Motor
Motor 3200U/min -> max Speed 56000mm/min -> 933.3mm/s
P1 -44 -> 1
P1 -45 -> 1
P1 -46 -> 4375
Motor Tuning
steps per -> 1000
without and electronic gearbox.
btw, what is your paramater P1-00
sorry for the confusing.
-
Hi Thomas,
yeah those numbers would work too. I am beginning to think that OP doesn't have an ESS at all but one of those HM4...whatevers and it can't keep up.
There may be a set of parameters which would work, I suspect you'd lose a lot of resolution though. Seems a bit silly to spend big bucks getting a 1hp
servo and a good one at that, way better than the cheap run of the mill Chinese crap servos, only to get a Chinese control board that 'couldn't pull the skin off a
rice pudding'.
Craig
-
Hello Craig,
you are right that a couple of different parameter Setting should work.
and one ore the other would have some disadvantage.
but at the Moment stage (1 or 2 rev per min), ther must be an other
basic issue to restrain the Motor to spin in a expeted rpm.
Thomas
-
Hi Thomas,
the key to understanding was OPs last post where he described the servo speeding up as he increased the 'steps per' but up to a 'ceiling'.
That tells me that the controller/BoB is not signaling or capable of signaling at a respectable rate. My set of figures requires 114kHz pulse rate and I suspect
his controller is maxing out at around 25kHz. A genuine ESS can go to 4MHz!
Craig
-
so the quick and easy check would be to hook a oszilloscope to the puls Input of the Controller.
-
Hi there
Well Craig, you are pointing at an interesting issue here, attached is a picture of my Motion card, and there isn't any other BOB involved
in my system, all I have is a lap top with Win7, an ADAA - B2 Driver, 750 delta motor, Motion Card HM-LAN-4axis - and a 24 V, 10 Amp Power supply, and the ac power to my Motor.
also how about my LAN Card and/or it's settings in my Lap Top? Do I need to do some settings in my Mach3
Regards
Jhonansaro
-
Hi,
that is not an Ethernet SmoothStepper, its a Chinese made motion controller and they've put that name on the outside to suggest similarity with a genuine
Ethernet SmoothStepper by Warp9 Technical Design.
Do you have any documentation with it?
I am now almost certain that the controller is not up to the task of signaling your drive at 100kHz +. You have two alternatives alter your requirements such that
a slow controller will suffice or get a decent controller.
Craig
-
Hi Craig
Well that is a bad news, but good to know the cause of the problem, On my alternatives that you mentioned on my first one, what do I need to do? could you explain?
also can i add a BOB to my system in order to compensate for my short coming?
regards
jhonansaro
-
Hi there
On the issue of oscilloscope test, Could you explain briefly how can it be done? If you have clip on it would be great. my electrical ability is not my best part.
regards
-
Hi,
do you have an oscilloscope...good ones a very expensive, you would know!
Craig
-
Hi,
https://www.youtube.com/watch?v=1x3PvVUQEqo (https://www.youtube.com/watch?v=1x3PvVUQEqo)
Craig
-
Hi Craig
I was just wondering, that Mach3's Diagnostic Tab/Page doesn't have that Capability to check for frequency? It seems to me that It does have.
also on Mach3's Port&Pin Dialog it has a Encoder/MPG's Tab, How about the values of (Counts/U and Velocity), Do I need to modify these values.
I have not done a detail readings on these items as of yet.
I am really serious to get to the bottom of this issue, so I can go back to my Supplier in order to get refund or replace it to a better one.
regards
jhonansaro
-
Hi,
yes there is a pulse frequency in Mach Diagnostics but it relates to a parallel port which you aren't using, in fact you CANT use it, your PC and OS are wrong for it.
The frequency displayed then has no relevance to the output pulse frequency of the controller.
MPGs and Encoders are inputs to Mach, if you had a hand wheel for instance to jog an axis you would hook the hand wheel to the controller and then tell Mach how those
signals are to be interpreted. They have no bearing on the pulse output of the controller.
We need good documentation about your controller. I suspect that its Chinese and you'll get nothing from them. If you spent a couple of hours reading posts on the
forum from the last month alone you'd find 100 or more posts along the lines 'No support from Chinese manufacturer....No documentation....Doesn't work...'
and the list goes on. Unless you get GOOD documentation from them (small probability) then give it up as junk and buy yourself a decent controller.
Do not buy anything until you've sought and gotten some informed opinions on the board you are considering.
The Chinese must love us, we buy cheap, but 'That's OK they don't have to work and we're not going to respond to any complaints so we will take the stupid white mans money!'
Sorry...I get pretty worked up about the Chinese ripoffs out there. Of course Delta is Taiwanese/Chinese and they are great, if I could afford them I would be buying
flat stick. So don't assume that because its Chinese it must be crap but it true also that if its Chinese it may very well be crap or a ripoff. If you can't get positive
feedback from others BEFOR you buy don't touch it.
It may well be that your board is adequate but we can't get it to work because we cant interface it to your controller but could if they would only give us some
documentation and a circuit diagram of the output circuit. Fingers crossed.
https://warp9td.com/ (https://warp9td.com/)
This is what a real ESS looks like and a company who backs up what they sell. Of course I'm biased, I own and use an ESS....
Craig
-
Hi,
reminds me that the ESS needs at least one BoB, I use Homan Designs (Australia) BoBs, fast and good quality and fairly local to me. only 1500nm!
Craig
-
Hi,
I'm on my laptop so don't have a probe hooked up, maybe you can test it out.
function autozero()
local inst = mc.mcGetInstance()
local probeHand, rc = mc.mcSignalGetHandle(inst,mc.ISIG_PROBE)
local probeState=mc.mcSignalGetState(probeHand)
if probeState==1 then
wx.wxMessageBox('Probe already active...cannot proceed')
return
end
mc.mcCntlGcodeExecuteWait(inst, 'G01 G90 G31 Z-4 F40');
local zProbeStrikePos = mc.mcCntlGetPoundVar(inst, 5063);
mc.mcAxisSetPos(inst, 2, 0.060);
mc.mcCntlGcodeExecuteWait(inst, 'G00 G91 Z1'); --Rapid move to 1 inch above current pos
probeState=mc.mcSignalGetState(probeHand)
if probeState == 1 then
wx.wxMessageBox("Probe is still activated! Check to make sure probe is not damaged or still contacting strike plate.");
else
mc.mcCntlSetLastError(inst, "Z Axis now referenced.");
end
end
if mc.mcInEditor()==1 then
autozero()
end
Couple of things I should point out. You see I have written this as a function and looks very much like a macro. I do this when I'm writing new code so I can put
it in a separate folder within my profile and I can run it with the debugger in the editor. Having it in a separate file means I don't pollute my macro folder with
buggy code. Once I've manually stepped through it and run a few tests then I will put in into its final location.
Here again I follow a somewhat iidiosyncratic policy of putting the function in the screen load script, usually near the top. Then I will put a function call on the button, not
the script itself. It means that when I go to find the script and edit it or have to delete it because its that buggy I can find it easily. If I were to attach the script
direct to the button the script will be buried in the screen script somewhere....I find it so much more convenient to put it at the top of the screen load script.
If it works out really well by all means put it direct on the button. This script is for my purposes only temporary to help you out so when we're done with it I'll delete
it knowing exactly where it will be.
Let us know if it works...
Craig
-
Hi,
my apologies everyone, this is obviously posted it the wrong thread.
Craig
-
Hi there
I have not been able to go back to my supplier, I will get hold of them tomorrow.
by the way I was wondering that I think this pulse frequency starts from my lap top, Please correct me if I am wrong?
If that is the case and It basically Generates fro my Lap Top and Mach3, Could it be the case that my lap top is incapable of producing high frequencies?
Regards
Jhonansaro
-
Dear Craig
I came across some reading which I am attaching it for your viewing, apparently this documents shows how to calculate the Gear Ratio, which i found similar Info. in
Delta's Manual. The calculation of the gear ratio are as follow:
Travel 175 mm
Reducer 10:1
Pulse /Revolution 17 Bit 160,000
Desired accuracy .001 mm
Gear Ratio = 160,000/(175/.001)*10/1=10 off course it is 9.41
Please let me know your opinion
regards
Jhonansaro
-
Hi,
If that is the case and It basically Generates fro my Lap Top and Mach3, Could it be the case that my lap top is incapable of producing high frequencies?
No this is incorrect. Your laptop produces numeric data of the planned trajectory, position and velocity over time, PVT. The motion controller then generates the
pulse streams.
Travel 175 mm
Reducer 10:1
Pulse /Revolution 17 Bit 160,000
Desired accuracy .001 mm
Gear Ratio = 160,000/(175/.001)*10/1=10 off course it is 9.41
Please let me know your opinion
Yes that set of numbers would work, as would the numbers I've advocated and the numbers that Thomas came up with. In fact there are many combinations which would work.
At this stage it is not the servo and drive that is giving the grief, its the controller apparently unable to produce rapid pulse streams and without good documentation
we cant tell if its faulty or just beyond its design maximum.
I personally think you'll be required to get another controller.....you can do virtually nothing with this one without manufacturers support/documentation. If they had it
surely they would have put it with the device.
Craig
-
i personaly agry with Craig.
we are on the wrong side of searchihg the fault.
the servo drive is absolutlty doinig what it's suposed to do.
the problem is the the controler.
-
Hi,
if I had the controller I might be able to do something with it, I have an oscilloscope (three acatually, two mine and the boss one) and have used them extensively for
years. What I would be looking to do would be to characterize the output circuit of the pin for which you want to run the step pulse.
Is it open collector? Will it withstand 24V? What current will it sink? How fast can it signal when sinking 5mA say?
These questions should really be answered by the manufacturer. If you had a look a look at the Warp9 website you will find page after page on the ESS, input characterization,
output characterization, communication, setup and more, much much more. Likewise you'd find very detailed documents on the CNCDrive, PoKeys, PMDX and Vital Systems
websites. They all maintain a board on the Third Party section of this forum as well as having there own forums plus technical backup via email.
These companies have been suppling the Mach market for years and are good at it.
You need to lean on the supplier of your board. If they are unwilling to help ditch them and NEVER spend a cent with them EVER again. Its only when we as customers
boycott these a-holes that they'll get the message.
Craig
-
Hi there
Yes possibly there are a lot of scenarios that could work, but from the equation that I provided to you if you choose to set P1-44 ,and P1-45 both to an equal Number,
Then you are forcing the servo to run with an accuracy of .0001 mm, which is kind of Unreal.
Travel 175 mm
Reducer 10:1
Pulse /Revolution 17 Bit 160,000
Desired accuracy .001 mm
Gear Ratio = 160,000/(175/.001)*10/1=10 off course it is 9.41
Regards
jhonansaro
-
Hi,
the catch is that with an encoder count of 160,000 and no electronic gearing, that is numerator equal denominator then at 685 rpm the pulse rate
would be 1.82 MHz which is totally unrealistic. The evidence you have is that your controller can handle about 18kHz, a factor of 100 too slow!
Craig
-
Hi,
just re-reading your post, you may note that the default encoder is 2500 ppr or count of 10,000 per rev.
You could program it to 40,000 ppr or 160,000 count per rev but why? You gain a whole bunch of resolution
which you can't use. Don't bother with it....an encoder count of 10,000 gives you plenty as is and should be in
the sweet spot for signaling if only the controller would play ball.
Has the supplier of the controller come back to you yet?
Craig
-
You could program it to 40,000 ppr or 160,000 count per rev but why?
just to see, that it is also not working.
-
Hi,
still trying to digest the details but that encoder count doesn't sound right.
A 17 bit encoder I think has a count of 131,072 per rev. Double and triple check the manual
about that count number.
Craig
Sorry for digging up an old thread.
I'm getting ready buy a set of these, and was confused about the 17 bit encoder count. Yes, what I've read tells me it should be 131,072, but everything I see tells me it's 160,000. Including the spec sheet from Delta.
I guess they use "17 bit" as a ballpark value? It's actually a little higher, but much less than 18 bit would be???
Also, I think that P1-46 is a configurable encoder count output, not the actual encoder count used by the drive.
-
Hi Gerry,
I recall being confused by this at the time this thread was current.
I seem to recall coming to the conclusion that there is 131,072 lines and therefore with quadrature 524288
counts per rev. In effect the encoder count is synthesized from this potential. If memory serves the default
configuration is 10,000 count per rev. So for all the extra resolution which the servo is capable is discarded,
which is still 2.16 arc min.....plenty fine enough for me.
The problem that OP had here, aside from not understanding electronic gearing, was that he was using a 'Canadian
made SmoothStepper'.......bulls*********t.......a Chinese made rip-off of a SmoothStepper. I recall being bemused that this
guy had spent big dollars to get four 750W servos and drives, huge overkill for a hobby machine, and yet wouldn't spend
$180 to get a genuine Ethernet SmoothStepper to actually get the best out of them.
Craig
-
I seem to recall coming to the conclusion that there is 131,072 lines and therefore with quadrature 524288
counts per rev. In effect the encoder count is synthesized from this potential. If memory serves the default
configuration is 10,000 count per rev. So for all the extra resolution which the servo is capable is discarded,
which is still 2.16 arc min.....plenty fine enough for me.
I think your mistaken, as it clearly says on the image I posted 160,000 ppr.
-
Hi,
well how do you explain it then?
Craig
-
Hi Gerry,
I discussed this in reply #7 of this thread.
The parameter P1-46 allows you to set the lines per rev, the range is 4 to 40,000. At the upper end, ie 40,000 lines
equates to 160,000 counts because of quadrature.
Craig
-
I think that P1-46 is to configure a digital output, not the actual encoder count used by the drive.
I found a quick start guide that states that there are 160,000 counts. If P1-46 defaults to 2500, then this guide would be incorrect if P1-46 actually set the encoder count that the drive uses.
https://www.damencnc.com/userdata/file/1559_ASDA-B2_Quick_Start_Guide.pdf
If you could change the resolution with P1-46, then why would you need electronic gearing?
I'll have to do some testing when I get one.
-
Hi,
If P1-46 defaults to 2500, then this guide would be incorrect if P1-46 actually set the encoder count that the drive uses.
That is exactly the same train of thought that I went down......but as you see P1-46 does default to 2500 for an effective
10000 count encoder. Have a look at the manual that OP posted, my reading of leads me to conclude the P1-46 sets the
effective count of the encoder, not just the digital outputs, but the primary encoder loop. Pretty kool!!
Craig
-
I don't think so, but testing will be required to prove otherwise.
-
Hi,
page 203 Chapter7
Craig
-
Is that not saying how to set up for the output signal of the encoder, ie for feedback use?
Just had a quick squint at the manual and it seems to say
1.the encoder is 160,000ppr.
2.you can set the electronic gearing so as to allow your step inputs to be suitable for your controller.
3. you can set the encoder output so that feedback is of a suitable range for your control, if it requires feedback, think CSMIO/IP-A
I may be totally wrong as I have only had a quick look but that is what it seems to be saying to me.
-
Hi,
1.the encoder is 160,000ppr.
2.you can set the electronic gearing so as to allow your step inputs to be suitable for your controller.
That was my initial impression also....but the more I read I came to the opposite conclusion, namely that you can
program the encoder to have anywhere between 4 and 40,000 lines. It is the synthesized encoder that is used as the
primary feedback and electronic gearing is applied to that....not the 160,000ppr.
Craig
-
I suppose the only way we will know for sure is for Gerry or someone else who has one, to do some experimenting and post the results.
The Chinese do things in a strange way to our eyes but they are equally valid and may have benefits, the setting of steps per unit in the SZGH control I have on the lathe being one, it took me quite a while to get my head around what they were meaning but once I had things hooked up it became clear as I could see the results :)
-
Hi All,
I have taken delivery in the last couple of days of a brand new Delta 400W B2 series servo and drive.
I can confirm that the native and raw encoder count is 160,000 count (40,000 lines) per rev and that the electronic
gear ratio applies to that 160,000 number.
The manual shows something which would suggest this is not so, see attached.
On the other hand the software tuning page suggests exactly as Gerry and Hood have contended.
Experimentation has proved conclusively that the gear ratio applies to the raw 160,000 count encoder.
Delta is a Taiwanese brand and the manual is in 'Chinglish' but is pretty good....as "Chinglish' goes.
Like any modern servo its going to take a great deal of experimenting to explore all the tuning features.
I was of the opinion that I would have to use a 3:1 or 5:1 gearbox to match my existing steppers but the stall
torque on these small servos has to be experienced to be believed. I'm going to try direct coupling them instead,
they certainly feel like they have the grunt to match my steppers WITHOUT a reduction box.
Craig
-
Hi,
I've started a new mill project, much bigger and more powerful than my mini-mill. I have some great 32mm diameter, 5mm
pitch, C5 double-nut ground ballscrews for the X,Y and Z axes.
I decided to get three new 750W Delta servos, B2 series. I got two plain servos/drives and one braked servo/drive for the Z
axis. I took delivery of them a few days ago but ran into a hitch....operator error.
In particular as I posted earlier the electronic gearing applies to to the 160,000 count encoder. What I wanted to do was
have 16000 as numerator and 100 as denominator. These numbers were for me to experiment with the servos using a signal
generator as the pulse source.
f2=f1 x16000 / 100
Thus if I feed the servo a 1kHz signal from my signal generator it would, after the electronic gearing generate an f2
of 1600kHz and thus spin the rotor at 10 revs per second or 600rpm......simple right!
I programmed P1-44 and P1-45 accordingly....or so I thought. I was using the buttons on the drive to set the parameters.
The range of numbers that can legally be used as numerator/denominator are very large, up to 10 decimal digits. The display
on the front of the drive is only five digits wide. Thus you can only see part of the particular parameter without hitting
<shift> to observe the higher order digits. What I had done , accidentally was erroneously programmed 116000 instead
of 16000 as numerator. It was not apparent that I had made a mistake because the higher order digit was obscured from
my view.
This lead to several hours frustration before I tumbled to my mistake.
What helped in the end was that I had, as a separate purchase bought a USB-to-IEEE1390 cable that I could program
the Delta drives with the manufacturer software. When I finally got the software installed and got the right drivers installed on
my PC and was able to read the programmed state of the servo the mistake I had made with the numerator became obvious.
Indeed programming the drive is just so much easier with the software than trying to push little teeny-tiny buttons
that I would have to recommend that you consider a programming cable an essential item if you buy a Delta servo.
The USB-to-IEEE1390 that I bought was manufactured by Delta and cost me $62 including shipping.
RS232-to-IEEE1390 can be had for less, about $30. Of course you only need one cable, you can program multiple
drives in succession, no need for a cable for each.
Aside from the great assistance the software gives you to program the drive there is a comprehensive scope and monitoring
suite that allows very detailed tuning analysis. All in all the manufacturer software and a cable to hook your PC to the
drive is all but essential in my opinion.
Craig
-
Hi,
I've started a new mill project, much bigger and more powerful than my mini-mill. I have some great 32mm diameter, 5mm
pitch, C5 double-nut ground ballscrews for the X,Y and Z axes.
I decided to get three new 750W Delta servos, B2 series. I got two plain servos/drives and one braked servo/drive for the Z
axis. I took delivery of them a few days ago but ran into a hitch....operator error.
What is the accuracy your are aiming at ? Even with 1micro meter (1um) I assume you are unable to hit limit of those delta encoders (cpr 160,000 ?) as your maximum pulse rate is 4Mhz ?
my math:
32mm ball screw diameter / 1um of CNC machine accuracy = 32,000 ppr,
encoder resolution = 32,000 * 4 = 128,000 cpr (below delta encoder spec)
Finally 128,000 cpr * (1800 rpm / 60) = 3 840 000 (imp / sec) so almost 4Mhz of pulse rate with 1800 rpm max
What is your view?
-
Hi,
my math:
32mm ball screw diameter / 1um of CNC machine accuracy = 32,000 ppr,
encoder resolution = 32,000 * 4 = 128,000 cpr (below delta encoder spec)
No, not quite.....the diameter of the screw is 32mm, the pitch is 5mm per rev. Thus for a resolution of 1um
I require an effective pulse rate of 5000 cpr, the numerator/denominator could be:
160,000=5000 x N/D
so N/D =32
Thus the choice of N and D is very wide, for instance if D=1 then N=32 OR
D=10 then N=320 OR
D=25 then N=800 and so on.
At 3000 rpm (50 rev per sec) the pulse rate is:
pulse rate=50 x 5000
=250kHz.
As you point out the ESS has a max rate of 4Mhz, so at 250kHz its cruising. 250kHz is a little faster than the max recommended
single ended (open collector) signaling rate but very comfortably within the 'low speed' differential signaling spec of the servo
drive of 500kHz.
I have some experience with an Allen Bradley servo which I wanted to signal at 466.66kHz. Its servo drive has a recommended
max differential signal rate of 500kHz and a single ended rate of 200kHz. I can assure you that while 466.66kHz is possible
its much easier to signal at a lower rate. Thus I reduced the angular resolution to 2000 cpr (from the encoder max of 8000cpr)
and got the signaling rate down to 116.66kHz which can be done comfortably with a single ended transistor.
Note that this servo is used as a direct drive spindle which gives a resolution of 10.8 arc min ( with 2000 cpr) from a max
resolution of 2.7 arc min (with 8000 cpr). I practice even 10.8 arc min is much greater than I require. Reducing the resolution
has made the whole thing easier to do and still exceed my actual requirements.
My existing mini-mill has a linear resolution of 1um and is very adequate. Thus I am quite happy to aim for 1um resolution
in this new build. I could, given the capacity of the servo encoder, have a resolution of 0.03125um ( 31nm!!!) but at the
expense of complexity of the signaling side of the design. I struggle to even measure 1um......let alone 30nm so why
bother?
The rated speed of the 750W B2 series servos is 3000 rpm. Direct coupled to a 5mm pitch ballscrew results in a max axis
speed of 15m/min which is a marked increase from my existing mill of 1.2m/min. I have experimented with pushing the
max speed parameter out to 5000rpm, the servos max speed (cf rated of 3000rpm) and they still work a treat.
Thus I could have rapids (G0's) of 25m/min in field weakened mode and full rated thrust for machining moves (G1's) of
15m/min while still maintaining 1um resolution and a max signaling rate of 416.66kHz, well within the 'low speed'
differential spec of the drives. As far as I'm concerned that is speed AND resolution beyond what I ever anticipated
I could achieve with a hobby machine. If I want to use the full acceleration potential of my existing mini-mill (I can
tune it to 1g accel easily enough) I have to tie the machine to the wall or the machine tries to dance all over the floor.
The practical limit for the machine is about 0.05 g without undue machine motion and still provides excellent tool path
following (at modest axis speeds).
These 750W servos offer at least a 50 fold improvement in potential over the existing stepper/reduction drives I currently
use. Thus with the X axis bed/ballscrew/rails/vice/workpiece weighing over 200kg if I tune the axis to 1g acceleration
(of which the servos are more than capable of) then I'm going to have to bolt this machine to a concrete floor just to stop
it lurching around threatening to crush any-one or any-thing in its path!
The speed, torque and resolution of these servos continues to exceed my expectations. 15 m/min G1's sounds fine on paper
but when you see the axis machining at 15m/min you realize just how much power and even how much potential damage
can be done.....its scary!! Fun scary.....but still scary.
Craig
-
Thank you for answer. I will study it a bit later however why only 5000cpr are required (seems low) ? Does delta encoder have any some special encoding (X4?). What is the calculation behind? Lastly if max pulse rate from ECC is 4Mhz then are you are limited by servo drive counting capabilities? Did you considered eg. Yaskawa or Argon drive which claims to support up to 4 MHz count rate for step/pulse
[Argon] https://granitedevices.com/wiki/Argon_specifications
[Yaskawa] http://www.e-mechatronics.com/download/datas/catalog/cheps80000061/cheps80000061d_3_0.pdf
Regards
-
Hi,
I will study it a bit later however why only 5000cpr are required (seems low) ?
If the effective cpr is 5000 and the distance traveled is 5mm ( the pitch of the screw) in one revolution:
resolution=5/5000
=0.001 mm
=1um
As I posted earlier 1um resolution is entirely adequate. I could have much more resolution but why?
Does delta encoder have any some special encoding (X4?).
Yes, in fact the B2 series encoder has 40,000 lines which (x4) means 160,000cpr. The numerator and denominator
of the electronic gear ratio are based on 160,000. No need to confuse yourself with lines vs counts per rev.
You can program the drive to output a simulated encoder of anywhere from 4 lines to 40,000 lines should you require an
encoder output stream, that does not affect the basic encoder built into the servo nor the numerator/denominator
of the gear ratio.
Lets imagine that you wanted the max resolution that the servo can achieve, in this case 160,000 cpr or
8.1 arc sec (31nm if direct coupled to a 5mm pitch screw) then at 3000 rpm the required step rate would be:
step rate=160,000 x 3000
=480 MHz
That is well in excess of the step rate of the ESS and you could never signal a servo drive down a cable at that
frequency anyway.
Note that the B2 series Delta servos can be signaled at 4Mhz in its highspeed differential mode. But still the question
is why bother? I have used my existing mini-mill for years with a 1um resolution and never had need for finer resolution
so why do I require more with this new mill......I don't. That decision has made the design of the signaling that much simpler.
I am not familiar with Argon but Yaskawa is of course justifiably famous.......and about twice to three times the price and quite
frankly, they are not anything like two to three times as good.
There is only one instance that I would consider Yaskawa an that is if I required Ethercat, Yaskawa has a market leading position
in Ethercat capable servos. Mach4 at this time is natively step/direction. There is a current development by Kingstar/Interval Zero
that has made Mach4 Ethercat capable but it is not yet mainstream. I have, at this time, no need to step up to Ethercat so
why bother with (expensive) Yaskawa servos?
Craig
-
Thank you for your time and answer. Learned a lot from it. I assume that your ball screw lead = pitch means they are one start type (not double or four start) that's why you need only 5000 cpr for 1um. Can you tell where did you bought them?
I thought that to calculate needed pulse rate we take into account not ppr but quadrature counts/rev (cpr, unless some encoding is in place)
so 3000rpm (50 RPS) with 1000 ppr encoder (4000 quadrature counts/rev) I need 12 Mhz step rate
Anyway I also do not see a point to buy expensive servo set if we can't use their all capabilities. However we can also see that Delta price comparing to Chinese servo's is bigger.
eg. price for 750W from delta is 364$
https://pl.aliexpress.com/item/32326080829.html?spm=a2g0o.productlist.0.0.52923c68IKfP6Q&algo_pvid=f62e7d55-b9fd-488b-8fc1-f67669d79a71&algo_expid=f62e7d55-b9fd-488b-8fc1-f67669d79a71-13&btsid=558db7fd-b9b1-4379-8b06-8365530e1541&ws_ab_test=searchweb0_0,searchweb201602_5,searchweb201603_52
while Lichuan 750W servo set price is 264 $
https://pl.aliexpress.com/item/32521880849.html?spm=a2g0o.productlist.0.0.50734680wdtcuv&algo_pvid=c0121e19-2367-4a38-b1ce-dc17b729bb2a&algo_expid=c0121e19-2367-4a38-b1ce-dc17b729bb2a-1&btsid=746ce6a0-008f-421e-82bd-1e958a8bbaee&ws_ab_test=searchweb0_0,searchweb201602_5,searchweb201603_52
Last servo is capable of 2500 ppr so with 5mm pitch we can get 2um accuracy. Isn't so bad so why bother for Delta?. What else did you consider to select them (eg. documentation, support) or some other params (motor inertia, maximum peak current (A), encoder frequency response or maybe how fast it can stop/change direction etc
-
Hi,
Delta is not the cheapest servo out there but it is the cheapest 'quality' servo.
I considered DMM (Canadian brand manufactured in China) and Delta (Taiwanese brand manufactured in China) and
ruled out the even cheaper Chinese brands.
I have good results with Delta VFDs.....and they have been in business for a long time whereas the cheapest Chinese brands are
newcomers to the market.
There is always a manufacturer somewhere who will make an item cheaper but of lesser quality. People who buy
on price alone are that manufacturers legitimate prey.
I tend to buy the best I can afford....often the cheapest IS the best I can afford but in the case of servos I wanted
better than the cheapest Chinese servos....there are quite a few posts on the forum about them, by no means all bad but often
there are shortcomings with them. One of the worst complaints is the atrocious 'Chinglish' used in the scanty user manuals.
Secondly many of the cheapest servos can only be programmed by button pushing on the drive whereas Delta (and DMM)
have software that resides on a PC and can be used to program the drive. You will see that in my previous post I now regard
that feature as essential. Modern servos have that many parameters, tuning options an modeling parameters that button
pushing is increasingly unrealistic.
I bought the ballscrews off Ebay from DYGlobal in Korea. He has become a favoured supplier having provided linear guides
and ballscrews for this new build and linear guides and various Vexta stepper motors/drives for my previous build mini-mill.
https://www.ebay.com/itm/THK-Double-nut-ground-Ball-Scr-BNFN3205-5-638mm-FK25-Precision-Z-axis-CNC-Router/201960959809?_trkparms=aid%3D111001%26algo%3DREC.SEED%26ao%3D1%26asc%3D20160727114228%26meid%3Dd16a24a45f9b46f5a5d6059dc4911ec3%26pid%3D100290%26rk%3D1%26rkt%3D4%26sd%3D201960959809%26itm%3D201960959809%26pmt%3D1%26noa%3D1%26pg%3D2060778&_trksid=p2060778.c100290.m3507 (https://www.ebay.com/itm/THK-Double-nut-ground-Ball-Scr-BNFN3205-5-638mm-FK25-Precision-Z-axis-CNC-Router/201960959809?_trkparms=aid%3D111001%26algo%3DREC.SEED%26ao%3D1%26asc%3D20160727114228%26meid%3Dd16a24a45f9b46f5a5d6059dc4911ec3%26pid%3D100290%26rk%3D1%26rkt%3D4%26sd%3D201960959809%26itm%3D201960959809%26pmt%3D1%26noa%3D1%26pg%3D2060778&_trksid=p2060778.c100290.m3507)
Craig
-
Hi,
Last servo is capable of 2500 ppr so with 5mm pitch we can get 2um accuracy
Re-reading this and I don't think you are correct.
The 2500ppr (lines) encoders used on cheap Chinese servos are all Tamagawa encoders or a Chinese made copy of a
Tamagawa. They are pretty damn good, especially for the money, such encoders ten to twenty years ago would have been
considered 'state of the art' and very expensive.
2500 ppr with quadrature becomes 10000 cpr. Thus a servo equipped with such an encoder has an angular resolution of:
angular resolution= 360 / 10000
=0.036 degrees
=2.16 minutes of arc
If such a servo is direct coupled to a single start 5mm pitch ballscrew like mine then its linear resolution is:
linear resolution= 5 / 10000
=0.0005mm
=0.5um
So.....yes, the current entry level 2500 ppr (10,000 cpr) encoders offer very good resolution at affordable prices. It is not
really necessary in my opinion to have even greater resolution in an encoder but ALL the leading servo makers are going that
way.
For instance the latest Delta A3 series servos are equipped with 24 bit absolute multiturn encoders with battery backup.
This means for example that you could use 16 bits (65536 cpr) of resolution for the angle within any given turn and 8 bits
(+128,-128 turns) to keep track of the number of complete turns from the nominal 'home' of the servo. Additionally the encoder
is absolute reading, that is as soon as it is powered up it knows where it is whereas the cheaper previous generation incremental
encoders (like mine.....sigh) require an initialization procedure before that are effective in determining servo angular position.
With a battery backup when you turn your machine off and the servo does not lose home reference, when you turn your machine
back on you don't need to home it as is common practice. Pretty 'groovy'.....but as nice as those features are I don't really
need them and don't want to pay the extra to get them. Maybe you do....if you do be sure to tell us all about how they work.
Craig