How slow is too slow for... 1

[cryomark]

...VFD driven motors?
I have been drafted for a project, already well on its way to FUBAR, where a compressor may need to run at a small fraction of its' normal throughput. Background novel follows.

A small experiment decided a year ago that they wanted to recover the helium being used for the cryogenic cooling, instead of venting to atm. At $3000/ 500L dewar/week, that was reasonable economics. The experiment has ~0 budget, so we rebuilt and donated an old 1 cylinder compressor to take their spent gas and send it to our main plants' dirty gas recovery system. We've since realized that the experiment has two distinct modes-cooldown, where the gas mass flow approaches 5 gm/sec and running, with mass flow of ~.5g/sec.
The compressor is pulley driven by a 7.5 hp, 1800rpm that starts and runs across the line and is rated at ~2.5 g/sec. So-during cooldown, a hefty portion of the gas gets vented to atm anyway, because of inadequate compressor capacity, and during running, the compressor is 90% shunted by a hastily installed valve, to prevent pulling vacuum on the suction side and introducing air contamination.
since we have several more of these 1 cylinder Corkin compressors on hand, I've been charged with getting a second one installed parallel to the first and giving the operators a way of maintaining suction pressure at ~1psig, under all conditions. Both compressors running full speed are suffucient for the cooldown load.But the running load is 1/10th the cooldown load and since the compressor throughput is linear w/speed, the motors would need to run ~180 rpm.
The initial proposal called for the purchase of a single 15 hp VFD to throttle back the speed of both compressor motors. This seems very slow to me and I wonder about the motors (ODP) ability to cool themselves.
My questions are:
Can AC induction motors run at fractional speeds reliably for extended periods? Is there a lower limit for standard duty motors?
What are the concerns beside heating?
If the pulley ratio were decreased from ~7:12 to say 5:14, how much faster can the motors be run than their 60 Hz speed, to achieve normal compressor throughput? 3150 rpm (105Hz)?

Strang but we've never used a VFD/motor to simply drive a load at variable speed. We have 30 or so 7.5 hp VFDs in service but they all operate in a continuous regenerative braking mode.

 

[busbar]

If you are looking for boilerplate, of possible application and currently available for no-charge download at nema.org… Application Guide For AC Adjustable Speed Drive Systems, ICS7-2000 and ICS7.1-2000.

 

[jraef]

The quick answer is, no you will not be able to run a standard AC motor at a 10:1 turn down without separate cooling of the motor. About the best you can expect is 6:1, so about 300rpm on your motor, but even that is pushing it and depends upon a lot of other factors. To be conservative and safe, I'd say 3:1 for an older design ODP motor on a contant torque application like yours. You might be able to find or scab together a cooling blower for your motors given that they are ODP, but I would not even begin to guess at air volume requirements. Another concern that I would investigate is the efficacy of compressor lubrication at reduced speeds. That is a very big problem with some compressor designs, especially recips that rely on crank splash or mechanical lube pumps etc.

I would go with a trial-and-error routine here given your budget constraints and your stated availability of spare compressors. Try it, but if the motor smokes, put on a bigger motor (in order to have more mass to handle the heat)or an inverter duty motor designed for this use. If the compressor smokes, you need to abandon the idea or find a budget.

Another approach might be to put the VFD on one motor running at a higher speed, and add in the other one as needed across-the-line or vith a second VFD.
Quando Omni Flunkus Moritati

 

[Lewish]

jraef, has said it well.
One additional thought would be to use some of the spent helium to cool the motor. Since it sounds like you are loosing some to atmosphere anyway. Vent it thru the motor. Helium is inert, so there would be no electrical problem that I can see.

 

[jraef]

Cool idea! (pun intended) Quando Omni Flunkus Moritati

 

[jbartos]

Suggestion: Beside noticeable changes in the motor parameters at lower speeds (efficiency, power factor); mechanical vibrations could increase at lower speeds.

 

[cryomark]

Thanks to all for your response.
jraef- the compressor lubrication issue may well end up as the killing factor for this cobble job.
lewish-Oxygen deficiency hazard regulations prohibit any intentional venting of cryo gasses in an inhabited area
plus the main idea is to recover all of the helium.

 

[Lewish]

I didn't realize that this was an "inhabited" area. If it is, then you are in violation anyhow as evidenced in the second paragraph of your novel "a hefty portion of the gas gets vented to atm anyway,".
So, why would using a very small amount for cooling matter?

 

[busbar]

In the US, Helium is viewed as a strategic resource 'mined' from Texas gas wells, so to varying degrees venting/wasting it is frowned upon.

 

[Lewish]

Yep, Helium City, Amarillo, Texas.
Next to the other GRAND canyon - Palo Duro Canyon.

 

[cryomark]

lewish-I think the straight line logic got lost in my background exposition. Sorry.
Cooldown mode delivers 5g/s; the single compressor can only take ~2.8g/s, thus the extra must be dumped to atm or the resulting pressure increase adds too much of a heat load to the system. By adding a second compressor, we hope to recover all the gas and maintain suction pressure around 1 psig
The current venting is through a parallel plate relief valve plumbed to a 4" inverted vent stack outside.

After discussing this with the ME and researching the compressors a bit more,we've tentatively decided to:
1. Change the pulley ratio on one compressor so it's full speed throughput is about half normal. This should allow the bypass valve to regulate suction P in the desired range with the only waste being electrical.
2. Add a 7.5 hp VFD to the second compressor. This will permit a gradual ramp up to maximum capacity, preventing the major P upsets starting across the line would produce.
We will also be able to regulate with this compressor's speed if necessary.
3. Operators have agreed to change the cooldown process so maximum mass flow is reduced, although this extends the cooldown period.
We will probably spec a new inverter duty motor with the VFD, even though the anticipated min speed is around 700rpm. Even though it has no budget, the experiment has been moving forward for 5 years. They must have a rock they can squeeze somewhere.