Hello Guest it is November 22, 2019, 04:36:20 AM

Author Topic: CS Labs CSMIO/P-A  (Read 17733 times)

0 Members and 1 Guest are viewing this topic.

Re: CS Labs CSMIO/P-A
« Reply #90 on: July 22, 2018, 07:11:04 AM »
Hi Craig

Thanks for the info. Definitely worth buying one at that price.

Allan/ Craig

Probably needs a new topic but would you have any idea where to start with the wiring on this spindle encoder? I can't see any information on it but has 8 wires coming out.

Also, I am trying to identify what is attached to the back of the spindle. It looks like some kind of air pump which is attached to a fork that moves in and out. I have never used it and I did ask Denford what it was a few years ago who didn't know. Please see images.

Cheers
Mick
Re: CS Labs CSMIO/P-A
« Reply #91 on: July 28, 2018, 09:22:51 AM »
Hi Allan/ Craig

The CSMIO has a max voltage input of 10v on the Analog I/0. I found channel 0 to be faulty putting out 10v continuously. The tacho for the motors puts out about 9v per 1000rpm. Does this mean until the tacho output voltage was adjusted to match the command voltage there was potentially approx 27v being fed into the CSMIO (At 3000rpm). Possibly causing the failure of Channel 0.

The only reason I ask is I am about to set up the Z axis and thinking maybe disconnect the Tacho wires from the CSMIO until the tacho output is matched to the command input?

Cheers
Mick     
Re: CS Labs CSMIO/P-A
« Reply #92 on: July 28, 2018, 09:28:57 AM »
Sorry meant disconnect tach from the drive. :D
Re: CS Labs CSMIO/P-A
« Reply #93 on: July 28, 2018, 12:21:24 PM »
Hi Mick

No, I don't think this is at all likely. The tacho and the command input are isolated from each other as both feed via resistors  into a summing junction of an op amp if I remember correctly. This summing junction is a low impedance input. The command input is further isolated by the input differential amplifier.

Allan
Re: CS Labs CSMIO/P-A
« Reply #94 on: July 29, 2018, 05:32:13 AM »
Thanks Allan

Appreciate your help again.

Mick
Re: CS Labs CSMIO/P-A
« Reply #95 on: July 30, 2018, 01:41:06 PM »
Hi Allan

Sorry to bother you again. I set up my Z axis and was working fine until today when the drive decided not to enable. Do you think its worth spending money on these drives as they are 30 years old? I have found a company that can supply some new 20amp analogue drives cheaper than it was to repair the Norwin ones a couple of years ago. They have checked the specs of the Norwin drives and Sem motors and advise they will work fine.
The drives do not do tacho feedback though. Is this a major issue? It's so frustrating doing a retrofit and sometimes wonder whether to just spend the money on a new machine.

Cheers
Mick     
Re: CS Labs CSMIO/P-A
« Reply #96 on: July 30, 2018, 05:37:08 PM »
Hi Mick

You are certainly having some bad luck.

As you know, both Craig and I are converts to ac servos, and without doubt that would be the ideal solution if money needs to be spent. But I appreciate that this would involve greater expense and more work, both electrical and mechanical,  than simply replacing the analogue drive.

As to the suggested analogue alternatives, I'm not sure that the lack of tacho feedback would matter greatly. The tacho integrates motor torque to provide velocity feedback which effectively makes the motor speed proportional to the voltage applied at the command input. An equivalent function could be provided within the new drive, for example by taking feedback from the armature voltage. The same function can be provided by the Kd term within CSMIO. Crudely speaking, all these options boil down to the same thing in the end. In practice, you will probably need some re-tuning of the PID loop but I cannot see why you should not expect an acceptable performance if the drives have a similar specification.    

Allan
« Last Edit: July 30, 2018, 05:39:45 PM by Fledermaus »
Re: CS Labs CSMIO/P-A
« Reply #97 on: July 31, 2018, 01:16:09 AM »
Hi Allan

I will have a read up on what the KD does in the CSMIO and try and get my head around it. Good to know there are options available without replacing the motors.
Thanks for your time again.

Cheers
Mick
Re: CS Labs CSMIO/P-A
« Reply #98 on: July 31, 2018, 03:18:06 AM »
Hi,
I made this same argument earlier in this thread, whether you have a tacho loop in addition to the CSMIO loop or just the CSMIO loop the
net result is the same. Clearly the PID coefficients will differ but both will achieve dynamic equivalence.

You are now really starting to pay for your choice to retain the existing servos. You paid a premium for an analogue controller to close the loop,
700GBP= 915USD. You are now looking at more cost to repair or replace your servo drive which takes the spend to just over $1000.
 If you had replaced the servos with AC servos and drives you could have used an open loop Step/Direction controller like a UC300
or an ESS for $150-$180. Two 750W DMM servos and drives are $1052, taking the total spend to about $1200.
New and modern AC servos/drives are so superior in so many ways.

I am of the opinion that spending more money on your existing servos or drives is a case of 'sending good money after bad'.

Craig
My wife left with my best friend...
     and I miss him!
Re: CS Labs CSMIO/P-A
« Reply #99 on: July 31, 2018, 02:50:21 PM »
Hi Craig

Thank for confirming that we can do without the feedback from the motors. I appreciate that you had said this before but please forgive me as I am not anywhere near up to speed with PID loops, servos systems etc. This is something I intend to get more familiar with as I believe you can fix anything if you know how it works.

As far as the CSMIO goes Hindsight is a wonderful thing. The lathe came from a Rolls Royce training facility and had only done 800 hours.

The CSMIO cost me £500 as I can reclaim the Vat and I have just had costings to repair the drive from Norwin for £75. Or I can replace the drives for £100 each.

So my spend so far is £575 or £700 with new drives.

The reasons I chose the CSMIO is because it is well respected by loads of very knowledgeable people on the forum and is distributed throughout the Uk by several CNC specialists.
It also has the ability to deal with threading which in my experience is difficult to get right with Mach 3.
Changing the servos would also mean that additional power supplies would be needed. Mounting plates would also have to be made and belts replaced. Then you have the issue of matching the new servos to the old ones as far as torque etc. Are the servos you've mentioned equivalent to what I would be taking off? 

If this is the wrong route I will replace the drives and Motors with new. Not something I want to do just yet but if I do, someone at some point will be looking for that particular servo motor and pay good money for it.   


Cheers
Mick