Can DC drives be run without a tacho input? 3

[edison123]

Don't the DC drives need a tacho input for speed & current control?

The client is running the motor without tacho input and with 'armature control' and is facing the problem of drive tripping on overcurrent.



Muthu

http://www.edison.co.in

 

[blacksea]

Speed control with armature (back-EMF) feedback must be properly tuned by PI-filter. Or drive has friction problem really.

 

[djs]

Usually not a problem. Most in-expensive DC drives use voltage control with current boost. Can get speed control to better than 1% using this method.

 

[djs]

I might also mention in most shunt wound DC motors the armature current is directly related to the motor torque. So if the motor is consuming too much current most likely there istoo large of a load.

 

[edison123]

blacksea - Thanks. Does the PI filter need to be tuned under no-load or under load? Unlike ac machines where frequency determines the speed straight away, dc machines have got two parameters that fix the speed. Without a tacho, I am wondering how a drive determines it is the right speed.

dis - Thanks. The client changed the motors of different makes (both have same armature voltage, current, speed & KW but different field voltage and current) and claims previous motor had no such issues for the same load. Could it be the drive tuning issues?

Muthu

http://www.edison.co.in

 

[Skogsgurra]

Never bothered about any "PI filter" in my whole life with DC drives. But, of course, one needs to adjust Proportional gain and Integral action in the speed controller part - and that is probably what one of the posters meant by "PI filter" - which is something entirely different.

The speed controller is, when you run without tach or encoder, the same thing as the EMF controller and you need an isolated EMF signal fed back to the EMF controller. That signal is usually galvanically isolated (use an isolation amplifier) or derived from the armature voltage by using a differential amplifier plus a load (armature current) adjustment to compensate for the load-dependent speed drop.

It is a trivial thing to set such a simple drive up, and if you ran the machine in your test bench with good results, it is the customer and/or his consultant that needs to do the actual adjustments. They should know how to. Or get help from outside.

My question is: Why was the motor changed in the first place? Was it because the drive failed? If that is the case, the drive should be fixed instead of the motor. A simple resistive load bank (use incandescent lamps) can be used to test the drive. And then - when drive is fixed - connect and tune (if necessary) the motor.

Gunnar Englund

http://www.gke.org

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

 

[edison123]

Thanks Gunnar. I was kinda hoping you would weigh in.

Yes, the motor was loaded up to 350 A (rated 2000 A) in our shop and black commutation was seen. Vibrations in all directions were less than 2 mm/sec peak.

The client changed out the motor since they wanted to send the running motor for a due overhaul. Talking with their engineers over phone, I realized their understanding of how dc machines work was not that good. I told them what you said - call the drive supplier to fix it since it's a drive issue and since they don't have the expertise. And to fix the damned tacho as the OEM intended. Haven't heard from them since. Fingers crossed.

Muthu

http://www.edison.co.in

 

[edison123]

I still feel iffy about variable speed motors not having tachos/encoders as a feedback control.

Muthu

http://www.edison.co.in

 

[singapore11singapore]

Hi,
If this is a new motor you must do the autotuning......standstill and rotating, which will automatically include the PID adjustment you are talking about. You cannot use the previous values..and you may need to de-tune the PID a bit, as it will be tuned only to suit the drive and not so much the application or the mechanics of the system....found this with Siemens 6RA24/70 anyway.
You can run DC without encoder, but needs to be tuned correctly...overcurrent as you are indicating is usually a sign of needing an autotune.

 

[lukin1977]

Most DC drives I know have 3 alternatives for speed feedback signal

1. Tacho
2. Encoder
3. Armature Voltage (taken from the drive itself)

You can use each depending on the speed accuracy you need, been Nº3 the less accurate and Nº2 the most

 

[lukin1977]

About the tunning. There are 2 different tunes to be done
1. Current controller. This has to be done with the motor stopped (sometimes it is need to mechanically stop the motor from spinning during tunning due to residual magnetism or series field). This tunning gets the armature resistance and armature inductance values

2. Speed Controller. This has to be done with no load and with the machine free to spin. This tunning gets the P and I gains for speed control

 

[djs]

You noted that the new motor has different field voltage and current. Is the new motor being supplied with the correct field voltage and current? If the new motor has a field rated at a higher voltage than the old one and was not changed, that could result in higher than normal armature current. This is a condition known as field weakening.

 

[edison123]

Thanks singapore and lukin for the tips.

dis - yes, the new motor is being fed right field current and voltage.

Muthu

http://www.edison.co.in

 

[edison123]

Just now talked with the client. ABB advised them (without even going to site to tune their drive) to put back the old motor with the current drive and told them the new motor needs an upgraded drive since the present one is 'obsolete'. I sense sales talk bs.

Muthu

http://www.edison.co.in

 

[Skogsgurra]

Yes, Muthu. Obsolete is a sales pitch more than a description of the state of the drive.

ABB started it, if I remember right, and Siemens and others followed. The idea is that equipment shall be serviceable (there shall be knowledge and spare pars available) so that the customer/user of equipment doesn't need to be lucky to find support. It is now getting ridiculous, with equipment being declared obsolete after ten years or so.

But, for critical equipment, where a still-stand costs a lot more than the equipment as such, it may be a good idea to upgrade every ten years or so. The MTBF thinking doesn't seem to be valid any more.

I have seen so many examples of "obsolete" equipment in the new sense of the word that I am beginning to accept this thinking. We old-timers are getting fewer and the new guys are a lot better at more complex things, like data networks and graphics. And more sophisticated tuning is available in all drives nowadays. So, get new equipment that the new guys can handle. There are still enough old equipment left, obsolete in the real sense or not, to create plenty of jobs and challenges for us dinosaurs.

Gunnar Englund

http://www.gke.org

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

 

[edison123]

Thanks Gunnar. Got the OEM design data for both the motors below.

Does the drive has to be tuned for any of the above parameters? For example, does GD² matter at all?

Muthu

http://www.edison.co.in

 

[lukin1977]

What is the drive`s output rated current?

 

[waross]

Gunnar, check me on this please;
It looks as if the resistance of the armature plus compensating plus inter-pole resistance of the new motor is about 15% less than the old motor.
If the drive is using the applied voltage, compensated by the current to derive the speed, that change in resistance may be causing a derived speed error which in turn is causing the over current. If you can find the compensation circuit and add a small amount of resistance this may prove the concept. The added resistance does not have to be exact, it just has to show a reduction in the overcurrent. If this is successful you can then look for the appropriate adjustment.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[edison123]

lukin - I don't have the drive details yet.

Bill - I have a very basic understanding on how drives work. So, not about to advise the client on how to proceed.

Muthu

http://www.edison.co.in

 

[waross]

Understood.
My work with DC drives was back in the day of MG sets, relays, resistor networks and Amplidynes. My question to Gunnar was based on a partly remembered understanding of how the drive looked at the motor parameters to determine speed and to limit overcurrent. Although the drives today bear almost no relationship to the drives of yesterday, The motor basics and the relationship between the currents and voltage drops across the commutating poles, compound poles and the armature remain.
My post was more a question to Gunnar to help my understanding, than a suggestion to you. Sorry for the misunderstanding. My bad.

Bill
--------------------
"Why not the best?"
Jimmy Carter