[Drassk]

I'm thinking of retrofitting my '94 4020 and I'm wondering if it's possible to use the existing axis amps on the machine instead of having to buy new motors/amps.

Does anyone know how the CPU talks to the amps? Are they just +- voltage or is there some more complex communication going on there? It'd be awfully nice to just replace the brains without having to change out any of the power components!

[Neal]

The CPU talks to the axis contoller cards (1010) and they talk to the amps that then in turn send the commands the motors. The motors talk back to the axis contorller cards and the contorller card then passes the data from the motors back to the amps. The CPU does not talk directly to the amps.
That is the reader's digest explanation!

Neal

[Drassk]

Hey Neal,

I should have been clearer with my terminology there, CPU was a bad term to use. I know what's going on in the cabinet so far as the control's workings go, but I was wondering what type of signal is sent between the axis controller cards and the amps. IE: Are the amps taking a 0-12v or a -12 to +12 volt signal or something else.

I'll be sending the resolver feedback back to my controller, and driving the amps directly from there so I'm curious as to what they accept as input. We'll end up hooking a scope to them anyhow, but I figured it'd be a good starting point if someone already knew how they're 'fed'.

[Neal]

The DC resolver operates on 3.6 volts and the feed back signal is 1.8 volts.

Neal

[Drassk]

That's some good stuff, Neal, thanks! I spent some time getting a ground-up explanation of amps from an electrical engineer today, so I've got a better idea how to make the amps do what I want now. Do you have a Fadal of your own at home?

Word to the wise: I've found that the least expensive, and highest quality, education to be had in technical matters is to spend a week web-searching and reading books and then offer an expert a meal somewhere nice to 'pick their brain'. I got three hours on FPGAs, microcontrollers, power transistors, amplifiers, signal noise, bandwidth and frequency considerations in PCB design...etc etc...from a brilliant electrical engineer for $30 total.

[Neal]

Drassk--
I do not have a machine. I was the senior applications/programming and tech support engineer at Fadal for 16 years until the new owner MAG IAS decided I was no longer needed.

Neal

[Drassk]

I know; you've helped me get my machine (incidentally 16 years old) up and running again on more than one occasion. Randy Cordes told me to always as for you when I called. Letting you go was proof that they'd given up on maintaining anything of the former company. If you had a machine, I would have sent you one of the controls when I'm done in exchange for some help reverse-engineering the auxiliary systems on the thing (way lube, ATC, etc).

The less stuff I need to change out, the better it is for me initally, and I'm hoping I'll be able to get the tool changer working without having to servo it (it's a Geneva), though I'll do that eventually just for speed. I have my table surveys, and I think the backlash settings can be extracted from the controller(?), so the ATC looks like the biggest hurdle.

[Neal]

FYI-- the survey values are stored on the 1010 axis controller cards and not on the CPU
. The CPU can read them and report them.

Neal

[Drassk]

Nice, I didn't know they could be retrieved once they were entered. I have the paperwork where they're written down.

Can the backlash values be read out, too? How do you access that stuff? I was thinking I'd need to measure the BL, as well.

[Drassk]

SVV! I bought my machine used, so the only manuals I've had access to were online. I did a bunch more digging today and I found another set of manuals which included a bunch of stuff I didn't have access to before...like the Axis Drive Systems Manual (aka: gold mine)

I'm getting more hopeful about getting this whole thing to work!

[Neal]

To retrieve the backlash type BL at Enter Next Command. The backlash table will display. If it does not then there is NO backlash entered into the system.
The code for the survey is SV.

Neal