Brother A-00 board upgrade

[grinders]

Hello. I have a 1999 Brother TC-22A. These are a great machine but the early models lack a few pretty standard capabilities like dnc drip feed and 3D arc functions like G18 and G19. After almost two years I finally found a decent deal on the parts needed to upgrade (nc-pcb and memory-pcb). I got a B52J090-4 motherboard and a B52J073-2 memory pcb with version 25 software. I'll be documenting the process here since I haven't seen a thread like it yet. I've taken pictures of every page in the machine control and I'll be calling Yamazen for a procedure to set all the things (I think) I need to do like reset home and toolchanger magazine, and Spindle orient. I'm sure I'm missing some stuff but this is just the beginning. It will take a few weeks because I only get sundays in the shop. Feel free to chime in if you have any tips or tricks!
Thanks

 

[BROTHERFRANK]

That's a great idea and very generous of you. I am sure there are more than a few here that will benefit from your experiences. The question about upgrading the control features on early (pre 2003) A00 control Brother machines seems to come up pretty often. G18 and G19 plane selection is not available on any A00 control version though. You will have X-Z and Y-Z arcs available as G102/103 and G202/203 codes for clarification.

 

[grinders]

That's a great idea and very generous of you. I am sure there are more than a few here that will benefit from your experiences. The question about upgrading the control features on early (pre 2003) A00 control Brother machines seems to come up pretty often. G18 and G19 plane selection is not available on any A00 control version though. You will have X-Z and Y-Z arcs available as G102/103 and G202/203 codes for clarification.

Thanks for the clarification. I honestly didn't know there was a difference between plane select and 3D arcs, good to know.
I got a PDF from Pete deal with the memory pcb version/parameter history and have been going over it. One thing I'm not sure about is it says at version 23 user parameter 5 changes from 2 to 3 digits, so you have to back up and re input all the parameters. Is there a way to load this data the same way you output all data?
Also what are the 8 physical switches labeled "SW1" on the NC-Pcb?
I will get my manuals on Sunday (only get 1 day a week with the machine) and see if I have the correct maintenance book. Thanks!

 

[grinders]

Ok so I figured out how to output/input all data successfully. It's in the maintenance screen of the ptr input where I loaded programs. Parameter write needs to be enabled in the I/O screen. Unfortunately I have every manual but the maintenance one so maybe I can find someone with the right book. On a side note version 21 is when the tape drive function was introduced. Looks like it's gonna happen next weekend!

 

[Pete Deal]

Grinders I am right behind you. Next month I am going to order the mem pcb, new spindle, and Y axis linear bearings. I would like to keep it mine going a few more years.

 

[grinders]

It's pretty exciting! I don't even mind splitting programs, but when you have 60 of them, at 2 mins each, that's 2 hours of just loading. Not to mention things cool off and everything fluctuates in those 2 minutes, especially on machines with a few miles. With a little luck, this will be a smooth transition!

 

[grinders]

Ok, I got the boards in. That was the easy part. Most of my input data did not work, kept getting transfer errors. I had a feeling I might have an issue so I took pictures of every screen in the control and printed them out. First, I matched the 8 SW1 switches from the old board to the new NC-Pcb and powered up. Then I had to change the machine to inch and match the communication settings. Then I loaded the M/C and User parameters from the old version. Then I went up the revision history changing what it called for up to the revision of the memory pcb I got (version 25). I'm left with two alarms now.
8204 parity(tool data)
6194 magazine no error
I'll be back on Friday to clear the alarms, double triple check everything, and hopefully be back to running and try out the tape run!

 

[Pete Deal]

Once mine lost it's mind and I had to reenter all the parameters by hand since I had not yet downloaded them. I think maybe I had the same thing with the tool data error and had to go and enter at least something in all the tool data fields.

 

[grinders]

Must have made a mistake somewhere. It will not save any tool data in inch mode. I can input, but it will disappear as soon as I leave the screen. Works normally in metric mode. Need to figure out how to reset the machine position too. Guess I'll try again next weekend.

 

[Pete Deal]

I got my memory board installed and I think it is all complete. I made a few mistakes along the way. The first thing to do is to back up all parameters (inch and metric). I backed up the inch parameters but not metric. Next install the new memory board (with power off of course) and then I think a good step is to format memory. Then reload the user parameters from backup (they are lost in the format). Then enter all the parameters from the upgrade sheet that Yamazen supplied. I had to redo some things a few times because I did not format memory and after entering all the new parameters still had a memory error. Because I did not back up the metric parameters the user parameters in metric mode are gone. I don't use the machine in this mode so I don't know that it will matter. I need to look to see if I have the original metric parameters. I have not tested it much but it seems to work.

 

[grinders]

The memory format is what I was told will solve my tool parity error. I was able to clear the magazine error by "clear start p data" in atc maintenance mode. Yamazen said it doesn't affect it if you don't have the metric parameters. To fix the machine position, they told me to hold release, press reset repeatedly about 10 times, then zero return. Supposedly the encoders will ignore position and go to zero as normal. Getting closer!

 

[grinders]

All finished! Now it's time to dig around a bit in the DNC settings. Tape mode worked as intended, but while testing it would randomly drop a character or 3 and alarm out. Happened twice before I had to call it a day. Have 1500 meters(?) of memory available after the format. Next is fixing a tool change clunk, but that's a different thread. Super happy with the possibilities with this machine now! I'll update once the tape mode is stable.

 

[Pete Deal]

Tool change clunk? Mine did that too on the first change after my upgrade. I did a lot off close looking at it and looking at the maintenance manual. The maint manual talks about a jig or something to use when changing out the tool magazine. I wondered if a grid shift was needed or something but after looking at it some more during a slow tool change it did not happen anymore.

 

[grinders]

The clunk is not related to the board swap, it was doing it since I got it. It's a swing arm style, not the drum so it's a bit different.
Took some slo motion video and everything lines up pretty good so i'm leaning towards an issue with the draw bar actuator. It's a lever arm with a roller bearing on the end. There's a gap between it and the actuator shaft(?) on top of the z casting. I'm sure Yamazen has seen this before, and they solved my previous issues with a 2 minute call.

 

[BROTHERFRANK]

The clunk is not related to the board swap, it was doing it since I got it. It's a swing arm style, not the drum so it's a bit different. Clunk - Streamable
Took some slo motion video and everything lines up pretty good so i'm leaning towards an issue with the draw bar actuator. It's a lever arm with a roller bearing on the end. There's a gap between it and the actuator shaft(?) on top of the z casting. I'm sure Yamazen has seen this before, and they solved my previous issues with a 2 minute call.

I recommend you take the cover off the ATC cam mechanism if you haven't already. There are quite a few cam folowers in there you should check and grease. I have seen a cam follower seize up and proceed to grind the cam flat which eventually threw off the timing of the actions. A costly and time consuming fix that could've been avoided.

 

[grinders]

This cover?
Imgur: The magic of the Internet

 

[BROTHERFRANK]

Yes. It's sheetmetal, about 16" x 16" roughly as I recall. The ATC arm shaft goes up into the cam.

 

[grinders]

I checked out all the cam followers and none were stuck or dry.
http://imgur.com/a/wdU98
There is a lot going on in there and nothing seems worn or obviously damaged. The dnc issue was equally uneventful, it mixed up data at the restart once the buffer is full. If I slow down the baud rate so the control is ahead of the pc, it works just fine. Might get a proper 232 cable and try again.

 

[Pete Deal]

I bet your handshaking is not working. I have been using the brocomm program to download programs to the machine prior to the mem board upgrade. It uses hardware handshaking. I had to use a desktop with a real serial port (not a usb adapter) to get it to work. Then this week I did drip feed a program using the same setup. I will figure out what the cable wiring is- I made my own according to the help file with the program as I recall.

 

[grinders]

Made my own cable and it didn’t fix the random error issue. It works perfectly when sending programs for memory mode. I’m using a pc with a pci serial port card and dnc server for the file transfer.
Imgur: The magic of the Internet
The errors don’t seem to be in the same spot each time but always seems to be some sort of double sending data. Anyone seen this before?