Inverter making a dc drive run faster

I recently installed a Reliance GV3000 AC Inverter on a machine. It took a few days to notice but when another similar machine (actually there may be 2 machines) with dc drives are running and you turn on this machine with the AC inverter they speed up. Voltage (240vac) stays constant but current increases by 1 amp on the incoming side of one of the dc drive machines. The AC line voltage is 240v, one dc drive is a Boston Gear Ratiotrol (the really noticeable machine that speeds up) and the other is a Reliance VS.

I have the line reactor in place on the incoming side of the AC inverter but it doesnt seem to make any difference.

I have been busy lately and I am not thinking straight on this. Any ideas on what I should be looking at? I may have to use an isolation transformer but with cost I am not sure thats practical.

Let me know what you think.

I assume there's no mechanical coupling between the machines that could be causing the speed change.

VFDs do introduce harmonic distortion into the AC waveform, and for a given speed command, DC drives gate their SCRs based on a perfect sine wave, so I think I can see how a distorted wave could affect the drive's DC output voltage. Maybe Dick DV will see this post and comment based on his experience.

Is the speed change causing so much of a problem that you have to do something about it? An isolation transformer should reduce the harmonic distortion from the VFD. Another possibility might be to add tach feedback to the DC drives. That would give you tighter speed control. I'd vote for the transformer on the basis of curing the cause rather than putting band-aids on the symptoms.

Hmmm! This is a curious set of circumstances. Haven't seen quite the same thing before.

The first thing I would want to eliminate is the possibility of noise coming in on the speed reference to the DC Drive. I would set the DC drive speed based on the keypad or an internally generated preset speed and then see if the DC motor speed varies when the GV3000 operates. If the DC speed holds constant, there are noise problems on the DC drive speed reference and it needs to be filtered or shielded to get rid of it. If the speed varies as before and there is no speed feedback (tach or encoder), then I would conclude that the SCR's are not commutating properly due to high frequency noise on the incoming AC power leads. Further isolation and filtering will be needed. You might want to check with MTE in Monominee Falls WI for a cheaper way to do this than an isolation transformer.

If you are using speed feedback on the DC system, the problem could be AC noise on the speed feedback lines (much more likely with tach feedback than with a digital encoder). Using the internal fixed speed as above, I would switch the system to armature feedback (no speed feedback) and see if the speed varies with GV3000 operation as before. If not, you've got noise on the speed feedback lines. If it still varies, proceed to the SCR commutation stage as discussed above.

Another economic consideration, Ron, is whether it is cheaper to fix the problem at the GV3000 or whether it is cheaper to fix the problem only at the points elsewhere in the network where trouble occurs. For example, if the GV3000 is 200hp and the DC drives are 2hp, fixing it at the DC drives is likely cheaper. On the other hand, if the GV3000 is 2hp and the DC drives are 200hp----well, you get the idea!

I'd like to know what you find out about this. You've got my curiosity stirred!

To offer more details, the DC drive and AC drive are running the same type machine. When I did the changeover from DC to AC the supplier up the HP from 10 to 15. The DC machine is 10HP. Personally I wont them to replace the Boston Gear Ratiotrol and Reliance VS MaxPak drive on the 2 problem machines with these new Reliance AC units, the cost is about $3000 each. Overall I think that would eliminate many issues that arise with these older DC drives especially maintenance (brushes, commutator etc), energy factors and most of all fire(s) from paper dust. The new AC motors are dust proof etc.

Of course that means there will probably be harmonics added to the line but that can be filter/suppressed at the panel at a low cost. This line is 240vac and the only suppressor I have is for a 480vac line so I havent tried that yet. I did put a line reactor on the incoming side of the AC drive but that didnt change anything.

The DC drives do not have keypads etc....old drives. The speed is controlled by the operator thru a potentiometer. No encoder or tach is used.

Just to be sure I stated the problem correctly. Lets say the DC drive is running then you turn on the AC drive, the dc drive will speed up. If you turn off the AC drive the DC drive will slow down. I havent checked what the difference is in RPM but it is a noticeable difference. I am attributing this to harmonics but not sure yet if it is in the control or commutating part of the DC drives.

I will keep plugging away till I determine what the problem is. Thanks for any info.

Ron, these older analog drives are passing from the scene but I still remember some things about them, especially the Reliance units since I essentially cut my teeth in the drive field on them.

I would suggest disconnecting the operator speed control pot and setting the min speed pot on the problem DC drive to some reasonable operating speed where you know the speed increase phenomenon occurs. Except that this time the speed is set by the min speed pot in the controller rather than the remote pot on the operator console.

Now run the DC drive and then start the GV3000. If the speed increases as before, you've got commutation problems due to incoming AC power noise. If the speed doesn't increase, its likely noise on the remote speed pot leads. You know what to do for that!

It's a little bit remarkable that these older DC drives don't have isolation transformers ahead of them. Back in the analog drive days, isolation transformers on DC drives were almost universal practice unless the job was being done on an excessively tight budget. Maybe that's what happened in your case.