Machsupport Forum
General CNC Chat => Show"N"Tell ( Your Machines) => Topic started by: Johann on June 05, 2013, 05:38:36 PM
-
Hi people!
I'm Johann from the Netherlands.
Bought an old BPi412 a while ago.
(http://www.swededemon.com/images/area34/images2/nc/412/412-1.jpg)
Machine was supposed to have lost it's parameters. Price was accordingly and some tools came with the machine.
After receiving the machine it turned out the parameters were still there but the Bosch servo amp had a problem.
(http://www.swededemon.com/images/area34/images2/nc/412/servo-amp.jpg)
An attempt to repair the amp didn't work out so I decided to replace the servo's.
To at least be able to use a mill I ran in to another BP, a 308, at an auction and also bought this fully functional machine. While working with the machine with a HH 2500 control I came to the conclusion the 2500 was a tad to limited for what I wanted to do. A fun machine and with CAM software things were possible but communication and also the machine's memory are slow and limited.
It got me thinking about the BPi412 with it's HH 355 control. After checking on servo drives it also turned out that the encoders make things difficult or expensive better said.
Eventually I decided to rebuild to Mach3. A bit of a gamble. Not much experience with Mach nor G-code but I thought it was nice challenge.
This forum has been a great help to get to that decision. Next question was about the setup. Which controller to use and which servo's. Eventually I decided to use a CSMIO/IP-S in combination with 750Watt Delta ASDA-B2 servo's.
Not having electrical diagrams of the machine made it an even bigger challenge.
While waiting for the Bosch amp repair I partially took the machine apart.
(http://www.swededemon.com/images/area34/images2/nc/412/412-4.jpg)
This to clean up the huge mess inside the machine and also to paint the inside of the machine white to make it appear a bit less depressing, a bit more light inside. I want to keep the outside as old looking as it is.
Also repaired the ATC,
Before,
(http://www.swededemon.com/images/area34/images2/nc/412/atc-3.jpg)
After,
(http://www.swededemon.com/images/area34/images2/nc/412/atc-4.jpg)
Welded a crack in the turret mount and replaced the bearings. Cleaned everything and added some paint and grease.
After receiving a Delta servo motor I made new mounting plates for the motors and ordered new pulleys and belts,
(http://www.swededemon.com/images/area34/images2/nc/412/x-motor.jpg)
Removed the Bosch cabinet from the machine and made a new placeholder for the Delta drives,
(http://www.swededemon.com/images/area34/images2/nc/412/drives.jpg)
About three weeks ago I ordered a CSMIO straight from CS-Labs in Poland. Also ordered an MPG unit with it. Removed most of the electrical parts from the machine. Managed to find a very bad quality diagram of the machine and discovered that I had removed to much from the cabinet.. ;D
After 5 days the CSMIO arrived and I started rewiring the cabinet. The Fanuc Spindle was a bit of a headache but eventually I managed to trace all the pins. Got the Fanuc working with the CSMIO. Spindle Orientation is also functional. Reused the ATC motor and used the original wiring with it. Basically managed to get all the OEM parts working. Only tested one drive so far because I'm still waiting for cables. Drive appears to work well. Needs some fine tuning when everything is in place.
Not being an electrical guy I think the cabinet looks pretty neat so far,
(http://www.swededemon.com/images/area34/images2/nc/412/mio2.jpg)
My biggest fear is the ATC. Did a lot of reading on this forum and decided to take a look at, the amazing, JHChoppers's script (http://www.machsupport.com/forum/index.php/topic,13365.msg112543.html#msg112543).
After having the pneumatics up and running I did some trial and error testing. I'm not a VB scripter by any means but after about an hour I managed to get the ATC to initialise and count pockets. Somehow the LED's of the Airblast, tool release and ram movement aren't working but the buttons are doing what they're supposed to do. Orient needs to be done. No clue about monitoring the Fanuc spindle though.
Just wanted to share and also thank JHChoppers for the great script.
-
Looks great and I agree, your wiring is very nice and neat :)
Personally I think the IP-A is the better choice but the IP-S is still an excellent controller and I am sure you will like it a lot.
Hood
-
Looks great and I agree, your wiring is very nice and neat :)
Personally I think the IP-A is the better choice but the IP-S is still an excellent controller and I am sure you will like it a lot.
Hood
The S has a bit more IO standard. Also CSLabs recommends the S for a modern setup. I already had the M so the choice for the S was pretty much obvious. So far it is working for me although I'm starting to run out of IO.
Have the machine running now from the computer screen. Still need to make the operator panel. Order a couple of buttons and a hand wheel. Started mounting a monitor to the panel,
(http://www.swededemon.com/images/area34/images2/nc/412/monitor1.jpg)
Not sure about the layout of the controls yet...
-
I am not sure why CS-Lab say that but from my point of view the A is the controller of choice. I have used the S and M on two other retrofits and both are excellent controllers, the A however has advantages over the S and M and will be what I use on the big lathe when threading can be done correctly (Mach issue at the moment but Brian is working on it :) )
With the Step/Dir models there is no encoder feedback to update Machs DROs, that means any time you take Mach out of reset you need to home as there is a good chance the axes have moved slightly from previous positions, that is especially true of a Z axis. When I press Reset on my Chiron the Z will fall slightly as it takes a few mS for the brake to hold, the X and Y may also move depending on forces acting on them. As the A updates Machs DROs from encoder values Mach always knows exactly where the axis is so when I take out of reset I do not have to re-home. Also moving an axis manually when servos are disabled will update Machs position.
Another nice thing the A has the advantage over the S for is the Index homing. The S can do it but depending on servo drives you may require additional hardware, the A you dont need anything extra.
The only disadvantage the A has compared to the S is less I/O but an extra I/O module would make up for that, obviously thats extra cost compared to the S.
So basically with the exception of the amount of I/O the A does everything the S does but it also does more which in my opinion makes it the better of the two.
As said however the S is an excellent controller and you will be delighted with it.
Hood
-
I believe you :)
But I had to make a decision and noticed this writing on CSLabs's site,
Step/dir control type has the advantage that the control signal is transmitted digitally and all motor control loops are in the drive, synchronized with each other, making this system work more precisely. However- analog control allows you to connect some old servo-drives, which can be a great advantage for people who want to retrofit their control system on old-type tool machine.
This and that I already had an IP-M made me decide to go for the S. With the M I only did some testing with a couple of steppers and started on a scrap build router using hybrid stepper drives but is a slow going project. I assumed Mach would use the controller in a similar way. Never knew the A will keep the position at all time.
(http://www.swededemon.com/images/area34/images2/nc/cnc.jpg)
Very far from finished though..
For the Z-axis I use the drive's output for the brake function. This because the drive knows when the motor is powered. The assumption here is that it will account for the latency. The only remark I have with this is that it appears the brake is operated more frequent compared to when operated by an external control.
Index has been set up as recommended by CSLabs.
-
My brake is also via my drive and yes you can configure the drive to apply the brake before the axis is disabled, however you will still have forces acting in that situation and when you remove the drives enable it will still likely move a small amount. The amount mine moves is small but it has still moved and thus each time you do that with the Step/Dir controllers the axis will be out of position. Also enabling the drive will likely move the axis slightly, and this is not just Z but all axes, again it will be a small amount but as Mach does not know anything about it you need to home after each and every time you take Mach out of reset. Its no big deal and it is something I do on all of my other machines that use SmoothStepper, Ethernet SmoothStepper and also the CSMIO/IP-S and M but it is nice not having to take that into consideration when using the IP-A controlled Chiron
Regarding the quote you attached from CS-Lab, well that can be true to a certain extent but there are a few other things to factor in. For example the drives you have are only closing the loop to their respective motors. With the IP-A the loop is closed in the control itself which is able to view ALL axis at the same time and I would imagine that would then allow the IP-A to then control all axis in sync with each other.
Again dont get me wrong, a step/dir setup with well tuned drives will be extremely precise, my big lathe and Beaver mill are but the IP-A is a step up from that from my experience.
Hood
-
Good work Johann. I like Delta servos. Their A2 series implement the technology used in the Mitsubishi J2S and J3 drives, for resonance suppression and continuous real time auto tuning. I am interested how you got the Fanuc VFD to work?
Dan
-
"Eventually I decided to rebuild to Mach3. A bit of a gamble. Not much experience with Mach nor G-code but I thought it was nice challenge."
Wow, if your doing this level of work, and not much experience, then I will LOVE to see your work, when you become experienced!!! Your work, is EXCELLENT, Great Job!!
I have never used the CSMIO/IP-x, series of controllers....... but the way Hood is going on and on about them, maybe I should try one!!
Again, really nice work!!
Scott
-
I am interested how you got the Fanuc VFD to work?
Dan
Biggest issue was that I didn't have info about the connector. Which wire did what. A huge puzzle.
The Fanuc uses 0-10v speed, CW and CCW direction, Alarm, Drive ready, Speed achieved and zero speed, Orient and orient achieved and E-stop.
"Eventually I decided to rebuild to Mach3. A bit of a gamble. Not much experience with Mach nor G-code but I thought it was nice challenge."
Wow, if your doing this level of work, and not much experience, then I will LOVE to see your work, when you become experienced!!! Your work, is EXCELLENT, Great Job!!
I have never used the CSMIO/IP-x, series of controllers....... but the way Hood is going on and on about them, maybe I should try one!!
Again, really nice work!!
Scott
Thank you Sir!
I only worked with Heidenhain dialog so far. Have been working with a 410 for a couple of years. The 308's 2500 is a bit older dialog and missing some cycles compared to the 410. Both can do G-code (ISO) but I never bothered. G-code is new for me. I think it has been 6-7 month since I started using, or testing better said, Mach3 and most has been configuring Mach itself. I have been working with computers all my live and learned an older programming language some 30 years ago so I'm not totally unfamiliar with the matter but a lot is forgotten over the years.. From the electronics side I do have some recent experience wiring cars, retrofitting standalone engine management systems in this case, and in the past I did maintenance on printing presses. So also here not totally unfamiliar. Never rebuild a wiring cabinet to this extend though. I did have some help but wanted to do the actual wiring myself.
A small video clip of the first run.
http://www.youtube.com/v/omeSv9P2cy0
(There a loud air compressor running in the background)
-
Hi Johann,
Would you share the Fanuc connector pinout?
Dan
-
Hi Johann,
Would you share the Fanuc connector pinout?
Dan
I'll post the info tomorrow.
-
Thank you!
Dan
-
Looks to be running great :)
What are the rapids on it, looks quite fast.
Hood
-
Dan,
Fanuc type is A06B-6055-H103 with a A20B-1001-0120 orient board.
Connections of CN1 are,
1,2 Speed achieved
3,4 Zero speed
7,8 Ready
11,12 Alarm
22,23 Orient achieved
25,26 Orient
31,32 Analog speed 31=0V 32=0-10V
14,45,46 Spindle start and direction 14=common, 45=FWD, 46=REV
47,48 E-Stop
Hood,
Rapids are set to 275IPM now. Accel needs some further adjustment still.
-
Thanks a lot Johann!
Why are the output pins in pairs? Are they relays or what?
Dan
-
Looked faster than that but thats still a very reasonable rapid :)
Are you set up in imperial units?
Hood
-
Thanks a lot Johann!
Why are the output pins in pairs? Are they relays or what?
Dan
They are contacts. With the BP the signals run from 24v. The inputs are NO or NC over a relay.
(http://www.swededemon.com/images/area34/images2/nc/412/fanuc-pins-1.jpg)
(http://www.swededemon.com/images/area34/images2/nc/412/fanuc-pins-2.jpg)
Looked faster than that but thats still a very reasonable rapid :)
Are you set up in imperial units?
Hood
I'm using metric. The speed isn't bad and OK for now although, from the info I can find, the BP412 specs show much higher speeds.
-
So how do you use the Orient pins, for example, do you have to short pins 25 and 26 to command Orient to the VFD?
And what happens, for example, when speed is achieved? Does an internal relay closes pins 1 and 2?
Dan
-
So how do you use the Orient pins, for example, do you have to short pins 25 and 26 to command Orient to the VFD?
And what happens, for example, when speed is achieved? Does an internal relay closes pins 1 and 2?
Dan
Yes when you connect 25 and 26 the spindle will position itself and when the position is reached the orient achieved pins 22 and 23 are closed. when you open 25-26 the orient procedure is stopped and 22-23 will open again. While 25-26 is closed the motor is powered and keeps the spindle locked in position.
This weekends effort,
Machined and sand(glass)blasted the operator panel,
(http://www.swededemon.com/images/area34/images2/nc/412/op-panel.jpg)
Hope to receive the buttons this week.
-
Yes I think likely 10 to 12m/min would be normal rapid for the BP, maybe even more.
Panel looks great, blasting has given it an excellent look, may have to invest/make one as that is impressive :)
Hood
-
Thanks Johann! This was very helpful info!
I like the panel too. Sand blasting finish gives it a good look.
Dan
-
It has been a while..
Have the machine running for a couple of months now but only started to machine some more parts with it for about a month now.
Putting the machine back together took some time. Decided not to touch the outside but the inside was painted white with an industrial coating.
Here the back panel can be seen,
(http://www.swededemon.com/images/area34/images2/nc/412/rear-panels.jpg)
This is what it looks like now,
(http://www.swededemon.com/images/area34/images2/nc/412/bodywork3.jpg)
Still need to make a better arrangement for the controls and keyboard but it is functional.
Started to make some parts,
(http://www.swededemon.com/images/area34/images2/nc/412/mani-top.jpg)
(http://www.swededemon.com/images/area34/images2/parts/3sge-intake.jpg)
And trying to get to know Mach3 a bit better. Making good progress IMO.
-
Looking good and seems to be working very well, certainly a nice looking manifold you have made :)
Hood
-
I found an interact 412 1986 model with hidenhan control i'm seriously considering buying and retrofitting. I'm sure it works fine as is, but plotting, finding centers etc etc is just so easy with mach 3.
How much in the end do you think this retrofit cost? Hood as you said in another post i'm thinking the CSMIO A or S controller, but using the drives that are already in the machine unless i MUST buy new drives? Hood if decide to go ahead can you give us a hand to get organised and buy the right stuff. I'd struggle without your help and experience, but I'd certainly pass it on, I tend to be ********* with anything to do with electronics.
Thanks guys. Heres a link to the machine, what do you both think?
http://www.machines4u.com.au/view/advert/Bridge-port-for-sale/270610/
-
If keeping the original drives it would be the IP-A you would want, it is the analogue command one, the IP-S is Step/Dir and your drives will almost certainly not be Step/Dir.
As for price, well not really sure, you would definitely need the IP-A, you would also need the Enc module if wanting Rigid Tapping and likely you would also want the MPG module so that you can easily integrate the MPG and buttons on your panel.
What else you would need on the machine I don't know, maybe need a spindle encoder, although there will likely be one there, apart from a computer then I think more or less what I have mentioned, plus time of course, should about do it.
Hood
-
I listed the IP-A with the Enc module in the bargain basement if you're. We can talk the price if you're interested.
Dan
-
Damnit this machine got sold fast... the hunt is on... soon as I find something I'll definitely get onto all of you... I will need help with wiring... luckily my house is insured for fire...
-
Probably someone saw you post the link here :D
Hood
-
That price caught my eye immediately.
It wouldn't have lasted long around here.
Be careful what / when you post, I doubt that you will get a commission for the sale. ;D