Yaskawa. Ah the memories. 1

[itsmoked]

Jeff, anybody?

I have a mill that has a Yaskawa spindle drive (MT3 3.7kW) that has, over the last year, started failing to go READY. This is because it's throwing the -ICAN error which is a 'memory error'.

I'm a little disappointed that the manual which lists about 5 different 'memory errors' doesn't specify what kind of memory error ICAN represents. Like, is this a forgetful EPROM issue? There are two 64k EPROMS together and one 8k off by itself. It would've been nice to know which group if that is the problem.

The error is thrown, dang near instantly, on power-up. About 5% of the time it starts up without the error and runs normally.


Considering:
1) Pulling the PROMS, reading them, and reprogramming new ones in the hope that my reader is able to get whatever value is fading.

2) Trying to find someone to supply a set of programmed ones.

3) Dumping this drive and replacing it with something like a CV7300 series TECO. (Spindle goes to about 6kRPM.)

4) Sending it in to be hosed by a repair shop ~ $1500

5) Buying a 'refurbished' unit off ePay.


4) and 5) seem to financially out of reach for the customer. He's going to have to switch to manual mode on an old mill and abandon in-place this mill.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

Keith,

I used to do a lot of Yaskawa drives back in the day.

This error is usually attributable to a dying power supply. The main board has a watchdog time on it and if the bus voltage comes up too slow, it gives you this error because it couldn't read the eprom in time.

Not sure how old that drive is, but if it's old enough, the PSU would be a smallish separate board. But I have heard the newer versions have the PSU built-in to the motherboard and it's all SMT. If that's the case, replaceing the entire board is probably your only option (but then again, it is YOU, so maybe not!).

Here is a procedure to try. If it recurrs, you move on to option 2.

Make a note of all the parameters.
Performing the following steps will re-initialize the Drive and return it to out of the box configuration.
1. Set the Drive Parameter Cn26 = AAAA
2. Power Off, wait 10 Min. Power On.
3. Set the Drive Parameter Cn25 back to what it was. (Motor Code, tells the drive what size it is)
4. Power Off, wait 10 Min. Power On.
Parameter Defaults are now loaded.
Re-enter the other parameters as above but change the data using the arrows and then store it by holding down the data key until it stops flashing.


"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[itsmoked]

Great. Hadn't considered that as an issue. Wish they had a reset button as that would let me 'reboot' with the supply already up. That would instantly prove that as the problem and let people limp by for a while too.

This drive is from 88" and is covered with about a hundred DIPs. Not an SMD part anywhere.

I check that out. Thanks a bunch.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[crthompson]

I'm not familiar with Yaskawa drives, but I have had internal "memory" batteries die and give me memory errors before... Something simple, cheap, and worth checking...

 

[itsmoked]

Hi crthompson.

I couldn't find any sign of a battery anywhere on that drive. I also didn't see any RAM on it and it's way too old to have EEPROM anywhere. I didn't scrutinize every one of the hundred chips though.

I ran Jeff's re-initialization recipe today and bingo! That seems to have fixed it.

Notably when the drive powers up it now plasters a large number on the display. Something like 50832 then goes to 0.0 instead of going instantly to -ICAN.

Prior to that I ran though all the parameters comparing them to a printed paper found under the drive from the factory. Any differences where duly noted. While many values were subtly different as expected one was very different.

On going thru the initialization process that value went back to a Yaskawa initial value. Something like 2.0 when I tried to set it back to the value I wrote down,8423, it wouldn't accept it and defaulted to the paper value.

Further power cycling always came up ready -something unseen in a week. As I write this it dawns on me that the only thing I didn't test was turning the machine off and waiting for 30 minutes to see if it forgets. I don't expect it to.

My current theory is that there is some sort of memory - volatile or non-volatile - that was corrupted (that crazy out of range number) and some sort of initial range checking was tripping out on it.

If this happens again I will look for a wild number and only correct it to see what transpires.

Thanks again Jeff.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

I'm just glad thaqt what my wife describes as my "fountain of useless information" managed to spew something that did some good.

Side note regarding the issue of errant high value numbers. We had a similar repeating failure mode at one time on our MV soft starters and it turned out to be a bad A/D converter. As it started it's death spiral, it would start doing this; injecting crazy high or crazy low values. The end result would always be the same I'm sorry to report; the eventual total failure. But if it's not an SMT drive as you say, you may be able to ID it and replace it if you feel so inclined. The bad ones that we found were all from Analog Devices, the more we investigated the more bad ones we found and ended up having a recall because of them. Turned out they were mis-certified for operating temperature. We thought we were ordering 60C devices, but they were 40C and that was way too low for our use. We had them on our currrent sensing circuit and when they went squirrelly, they would give the mP erroneous values of things like 52,486A of current, which would trigger the short circuit trip function (to say the least) and freak out the end users when they would read the fault history. That was fun...

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[itsmoked]

Noted...

And this thing does have an AD A/D on it...


On a side note if someday I need to pitch this thing and go with something else, what can you tell me about this spindle encoder? It has 8 wires; three-pairs, power, and ground. Does that tell us anything about what the encoder is?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

Jeff, that 'fountain of useless information' is what we engineers call experience. I think that Sheila needs to know that. Tell her.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[ScottyUK]

Keith,

At a guess for that encoder: quadrature squarewave outputs plus an index pulse on the third pair?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[itsmoked]

That's what I was thinking too. Now I need to see if replacement VFDs would generally understand that index pulse.

Thanks Scotty.

Keith Cress
kcress -

http://www.flaminsystems.com