Welcome, Guest. Please login or register.
Did you miss your activation email?
October 21, 2018, 09:09:13 PM

Login with username, password and session length
Search:     Advanced search
* Home Help Search Calendar Links Login Register
+  Machsupport Forum
|-+  Mach Discussion
| |-+  Mach SDK plugin questions and answers.
| | |-+  Finished Plugins for Download
| | | |-+  Huanyang VFD controller plugin
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 »   Go Down
Print
Author Topic: Huanyang VFD controller plugin  (Read 403892 times)
0 Members and 1 Guest are viewing this topic.
cncman172
Active Member

Offline Offline

Posts: 177


View Profile
« Reply #190 on: February 10, 2011, 10:29:08 PM »

I finally got around to testing this plug in on my Huanyang 3KW VFD.  Well after four hours I have not been able to get it to communicate properly.

Symptoms.

The VFD is programmed as follows:

PD001 = 2
PD002 = 2
PD163 = 1
PD164 = 1
PD165 = 3

Using the Chinese USB to RS485 that Mark and many others have successfully got working, setup as COM2 using 9600 N81

The plugin loads fine and I have NET3.5 loaded as well.  Plugin is activated and when I fopen the plug in setting and set it to COM2 9600 VFD 1 I never get the two parameters to populate like many others. I setup a serial monitor and can see the commands go to the VFD and it responds immediately with7F 7F BF to both requests.  Another symptom when I setup the VFD after power on the display flashes all "0000" the Power light is on and the forward light flashes and the ROTT light is ON. I have a shielded cable to the VFD for control linkage and the VFD and spindle are 7 feet from the MACH3 computer.  I even installed and EMI filter on the power input to the VFD, none of this made any difference at all.

Any suggestions as I am scratching my head.  I am usually pretty good at troubleshooting these issues but not this time.  The VFD and SPINDLE work perfect if I program the VFD for operator panel interface.  The RS485 does not respond at all if I reverse the leads going to the RS485 unit.  Thanks

Russ


Logged
cncman172
Active Member

Offline Offline

Posts: 177


View Profile
« Reply #191 on: February 11, 2011, 10:50:28 AM »

Just a follow up note with a little more information.
I tried to put 120 ohm resistors across both ends of the RS485, which made no difference at all.

When I populated the Huanyang plugin dialog box and hit ok it disappears, so i do it again to see if the PD001 and PD002 have changed from 99 to 2, but no such luck.  I am not sure that is the correct procedure.

Using the serial monitor program you can see the plugin sends out two strings and gets two responses.

Request: 01 01 02 01 00 B8 6C
Answer: 7F 7F BF

Request: 01 01 02 02 00 B8 9C
Answer: 7F 7F BF

I also read up on RS485 attempting to understand potential problems.  When I attempt to measure the voltage on the RS485 with respect to ground they are extremely low.  One is maybe 0.35V and the other might be -0.24V, which seems to be under the minmums listed for RS485. 

Any recommendations guys?  Thanks

Russ
Logged
majorstrain
Active Member

Offline Offline

Posts: 28


View Profile
« Reply #192 on: February 11, 2011, 10:57:46 AM »

Sorry Russ,
I'm no help at this time, I just ordered all the missing bit for the RS485 and braking functions off E-bay.
It will be a couple of weeks b4 they arrive and I get to the stage that you are up to.

Keep posting what you discover though, it will help myself and any others that run up against the same problems.

Cheers,
Phil
Logged
cncman172
Active Member

Offline Offline

Posts: 177


View Profile
« Reply #193 on: February 11, 2011, 11:22:28 AM »

Phil,

I plan to hook up a scope to see if the data is actually getting to the VFD.  Something seems to be going there because I am getting some kind of response.  I read from other others when you see 7F 7F BF as a response that is almost all 1's which usually indicates noise.  I have take so many steps to eliminate noise that seems strange.  The low voltage levels seem very strange to me, but if someone had one working that would be useful information.  I will hook up a scope this afternoon and see if I can actually see a data package actually getting to the VFD.  I might also open my VFD and look for other missing components.  I know that one relay near the terminal side of the board is missing and read about all kinds of other saying the MAXIM RS485 chip was missing.  Just unbelievable these guys ship this stuff with the missing parts.

I would also like someone to confirm the VFD should be flashing the "00000" the forward LED, and the ROTT led is on.  This might be normal but no one else has responded so I am pretty much in the dark.  Thanks for at least responding.  LOL

Russ
Logged
Sinij kot
Active Member

Offline Offline

Posts: 44


View Profile
« Reply #194 on: February 11, 2011, 12:26:40 PM »

Sorry Russ,
I'm no help at this time, I just ordered all the missing bit for the RS485 and braking functions off E-bay.
It will be a couple of weeks b4 they arrive and I get to the stage that you are up to.

Keep posting what you discover though, it will help myself and any others that run up against the same problems.

Cheers,
Phil

majorstrain, good job on the schematic.
I came to the same conclusion that the IGBT would be the same because the drive circuit is the same.

Where did you find a IGBT part for sale? got a link?

My board seems to have everything populated except for the Q23.

Thanks
Logged
cncman172
Active Member

Offline Offline

Posts: 177


View Profile
« Reply #195 on: February 11, 2011, 01:41:31 PM »

Another Update!

When I put an HP digital scope on the line leaving the USB to RS485 converter and tell the MACH3 plugin to bring up the plugin dialog box you can see two data requests fly across the link.  I checked both sideds of the line and they look good.  Seems like the data is getting to the VFD.  The VFD responds immediately two the two answers.  Not sure why it is not sending good data back.

I also opened the VFD to see if I had other missing components like so many have shared on the forum.  Well the only thing I see missing was one relay down near the connector with the screw terminals.  There is a spot for two relays I have one populated, which is  labeled as a JZC-33F 3A 30VDC.  Three contact pins C, NO, NC.  I saw several pictures on the forum of the boards inside the VFD.  The relay populated in mine is the relay near the outer edge of the board.  On other photos I see they populated the relay closer to the screw terminal strip.  Wow looks like a good deal of hand soldering took place inside the unit.  Clearly some design issues they added several capacitors after the fact which have not standard pads on the board.  They just soldered them between two other components.  I work in manufacturing and these boards would never pass our QA and we have boards built in China, Thailand, and Mexico.  Wow pretty surprising.

More testing is planned over the weekend, it would be great to get this thing to start talking to MACH.  Smiley

Comments, Suggestions, Welcome.  Smiley

Russ
Logged
majorstrain
Active Member

Offline Offline

Posts: 28


View Profile
« Reply #196 on: February 12, 2011, 06:05:22 AM »

Quote from: Sinij kot
Where did you find a IGBT part for sale? got a link?

My board seems to have everything populated except for the Q23.

I got the IGBT's from this seller http://stores.ebay.com.au/Asia-Engineer
The direct link is http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=390221440577&ssPageName=STRK:MEWNX:IT#ht_839wt_905
Got a lot of 5 for $3 a unit.

The opto's I got from this seller http://stores.ebay.com.au/wonderco-buy79
The direct link is http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=130416767452&ssPageName=STRK:MEWNX:IT#ht_1504wt_905
I bought 2 lots of 2 because the extra shipping was only $0.50

The 18V zeners I got from this seller in the USA http://stores.ebay.com.au/bigsmythe74
The direct link is http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=400059707817&ssPageName=STRK:MEWNX:IT#ht_500wt_1156
100 units for around $8

and lastly I got some RS485 chips from http://stores.ebay.com.au/yuqinguo47
The direct link is http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=320619204147&ssPageName=STRK:MEWNX:IT#ht_1191wt_1139
60 units for $15.

Just a note for Australian and NZ members.
If you want a couple of zenners and/or a Max485 chip, just PM me. I'm thinking if you send me a self addressed and stamped envelope I'll send it back to you with the parts inside.
Other countries could do the same thing but it would probably be cheaper buying them direct.


Quote from: cncman172
When I put an HP digital scope on the line leaving the USB to RS485 converter and tell the MACH3 plugin to bring up the plugin dialog box you can see two data requests fly across the link.  I checked both sideds of the line and they look good.  Seems like the data is getting to the VFD.  The VFD responds immediately two the two answers.  Not sure why it is not sending good data back.

I also opened the VFD to see if I had other missing components like so many have shared on the forum.  Well the only thing I see missing was one relay down near the connector with the screw terminals.  There is a spot for two relays I have one populated, which is  labeled as a JZC-33F 3A 30VDC.  Three contact pins C, NO, NC.  I saw several pictures on the forum of the boards inside the VFD.  The relay populated in mine is the relay near the outer edge of the board.  On other photos I see they populated the relay closer to the screw terminal strip.  Wow looks like a good deal of hand soldering took place inside the unit.  Clearly some design issues they added several capacitors after the fact which have not standard pads on the board.  They just soldered them between two other components.  I work in manufacturing and these boards would never pass our QA and we have boards built in China, Thailand, and Mexico.  Wow pretty surprising.

That's good news that all the required parts are there Russ. It might just come down to a setting issue.
Did you catch this post by stephenmhall on page 13, I'm not sure if it's relevant to your issue.
http://www.machsupport.com/forum/index.php/topic,14182.msg105213.html#msg105213

For the guys that have the diode D1 installed (under the bottom board), could you let me know the part number? I just want to confirm that it is a standard diode and not a zenner for some reason.

Cheers,
Phil
Logged
cncman172
Active Member

Offline Offline

Posts: 177


View Profile
« Reply #197 on: February 12, 2011, 08:56:02 AM »

Phil,

Hope your weather down under is better than we are having it up North.  It has been well below freezing for weeks, we are ready for spring.  LOL

OK, I did some more testing after reading your post and went back and read the other post "again" that you suggeted.  The only thing he did was change the VFD to #2

Well here are my results from both attempts:

VFD#1                                            VFD#2
request: 01 01 02 01 00 B8 6C           02 01 02 01 00 FC 6C
answer: 7F 8F BF                               BF FB FD
request: 01 01 02 00 B8 9C                02 01 02 02 00 FC 9C
answer: 7F 7F BF                               BF FB FB

Changing the VFD address in both places did change the output coming back from the VFD and looking at the data closely it almost seems like it is inverted.

BF FD in binary is 11011111 1111101 so if we invert it we get 00100000 00000010 and back to HEX is 20,02

I had a very lold Tektronics 338 logic analyzer which has not been used in probably 20 years and I could not find the user manual but was going to use it to read the output and input on the Texas Instrument 75176B the differential bus transceiver.
Pin 1 on the device is "R" the data coming from the PC in digital form, and pin 4 is "D" the data being send from the VFD back to the PC.  If I can get the logic analyzer or scope to capture this data at least I would know what the VFD is seeing and sending in digital format.  If this was correct then it could have something to do with the converter but right now I don't know.  I tried to play with the logic analyzer for an hour last night and started searching for a manual online to no avail.

More testing planned for today.  This has been very painful process.

Russ


Logged
cncman172
Active Member

Offline Offline

Posts: 177


View Profile
« Reply #198 on: February 12, 2011, 10:41:07 AM »

VFD Update: Well I spent two hour reading the 20 pages in this topic yet again, looking for some grain in data that I might have overlooked on how to establish communcations with the VFD.

I am using the 3KW VFD and Spindle which I purchased probably close to two years ago.  It works on the bench fine and just got busy with other projects so it never got installed on the mill.

I am using a USB to RS485 converter which uses the Prolific PL-2303HX chip, just like the one with photos and details at this link.

http://cgi.ebay.com/USB-RS-485-Converter-Adapter-RS485-New-17-/180612321384?pt=LH_DefaultDomain_0&hash=item2a0d555068

My cable between the converter and the VFD is connected in the following manner:
VFD RS-  >>> T+/B (DB9-2) and VFD RS+ >>> T-/A (DB9-1)
I have tried ground, no ground, and no difference.   One note when I measure the voltage on the TI 75176B differential chip on the VFD it measure 5VDC as expected, if I measure VCC to DCM on the VFD which is suppose to be DC Common I only get about 3.4VDC with respect to Pin 8  (vcc) of the 75176B chip.  However, if you measure between V0 and Pin 8 you get the full 5VDC.  DCM is not really the common DC side with respect to the IC's on the upper VFD pc board. Anyway I tried to use the V0 ground as a reference back to the PC usb RS485 converter again the same responses.

At this point I am not sure if the USB to RS485 converter I am using has been successfully used to control the VFD.  Hope this info helps someone else or highligths something that someone else has noticed that might point me to a solution.  I might blow another $30 on a different USB to RS485 converter.

I did noticed after reading these 20 pages of posts several times that Bryan "Bolingerbe" had the exact same issue and responces from his VFD but he disappeared and had not posted in a long time.  I know he was also frustrated and talked about get a different VFD fully modbus compliant.  I hate to go that route until this one dies because I already spent the money. 

Comments or Suggestions are welcome.  Thanks

Russ

Logged
cncman172
Active Member

Offline Offline

Posts: 177


View Profile
« Reply #199 on: February 13, 2011, 05:01:39 PM »

Yet another update!  I must be the only one attempting to get the RS485 running on the Huanyang VFD.

Well as a result of frustration I took the unit apart and pulled the top board containing the processor so I could trace some of the circuit.  I really wanted to see if data was getting past the 75176B differential bus transceiver.  After a good deal of circuit tracing I discovered that some of the key signals can be tested from on unpopulated connector CN4 a ten pin connector.  Pin 1 of this connector with the small white square is true ground with respect of the microprocessor and the 75176B.  Numbering counterclockwise from pin 1 you have the following key pins.  Pin 4 is the DATA OUT lead from the processor to the 75176B, this is held high via a 10K resistor R102.  Pins 5&6 tie directly to VCC (5vdc).  Pin 7 is READ DATA from the PC and goes directly to the 75176B and the data pin is held high via R101 a 10K resistor.  Finally, the enable leads on the 75176B are tied together and are held high via R40 a 10K resistor, and then attach to Pin 8 of CN4.

Attaching a HP 100Mhz digital scope to "R" the data pin coming from the PC RS485 I can see one 10uS pulse and that is it.  Nothing follows, which is strange as you would expect a good deal of data to come on this input pin to the processor in the VFD.  When I measure the Enable pins on the 75176B they stay low the entire time when I initiate the plugin.  This indicate the VFD never tries to write to the VFD.  I know this pin works because during a power cycle the pin starts high and then the processor takes it low. Again, this seems to indicate the VFD did not really get any kind of message it needed to respond to from the PC.

In addition if I put the scope on "D" the data output that comes to the 75176B this stays high the entire time again indicating that the VFD never attempts to send any data at all.  When I put the scope on the differential input pins A or B on the 75176B you can see a burst of data coming from the PC when I initiate the plugin.

What does all this mean?  Well I am now highly suspect of the 75176B, it was clearly hand soldered and I did reflow it using a precision repair station, but when data comes in the differential inputs it should translate to a stream of data on the R pin of the 75176B and I only see one 10uS low pulse.  Open to comments on my testing.

Another thing to keep in mind when it comes to the Free Serial Port Monitors.  While my scope clearly showed no data leaving the VFD processor the serial monitor was spitting out data.  My guess is this is nothing more than reflections of the data sent by the PC, could be all wet here but that is the only thing that makes any sense.  I will probably order a MAXIM replacement chip tomorrow and then seen what happens with a new chip. 
Hope these experiments help someone else.  If I can get time I will post a partial schematic on the RS485 interface on the VFD.  The interface looks traditional with pull up on the A pin (RS+) and a pull down on the B pin (RS-).

Russ Larson
Logged
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 »   Go Up
Print
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.20 | SMF © 2013, Simple Machines Valid XHTML 1.0! Valid CSS!