What limits the speed of an air compressor?

[strantor]

Is it heat buildup in the compressor head? Availability of lubricant to the crank? Other?

Let me explain the reason for my question. Compressors typically run off a simple pressure switch; on or off, 60Hz, fixed speed, fixed displacement. The CFM at 40psi and the CFM at 90 psi are usually pretty close, at least for those which use an induction motor.

For example a 5hp compressor, when you first turn it on, starts spinning at a more-or-less fixed speed, delivering a more-or-less fixed CFM, but consuming much less than 5hp. The amount of power consumed begins to climb as pressure climbs, and probably don't utilize all 5HP until right before the pressure switch drops out. The rest of the time before that, the true potential of that big motor is never utilized.

BUT if I were to put that motor on a VFD and (assuming motor is 3ph vector duty, with bearings rated for it) set the max frequency output to say 240Hz instead of 60Hz, and put the drive in torque mode instead of speed mode, then:
1. as soon as you turn on the compressor it ramps up to 4x the normal speed.
2. As pressure starts to build and the motor starts to see some resistance, it starts to slow down. This point at which it starts to slow down is the point at which it hits its 5hp limit, which it will stay at for the remainder of the cycle.
3. As the pressure continues to build, the speed continues to drop, until (assuming I've got all the parameters set correctly) it should fall just down to 60Hz right before the pressure switch cuts it off.

So the result is:
1. a much better utilization of the motor
2. A much faster charge-up time
3. Higher CFM (in fact much higher than nameplate rated) at lower pressures (and nameplate rated CFM at nameplate rated max pressure) instead of a more-or-less constant CFM.

Instead of a linear HP graph starting at zero and ending at 5hp right before pressure switch cutoff, we see a fast climb to 5hp and a plateau at 5hp, continuing until pressure switch cutoff.

So that's what leads me to my question. Is it safe to 4x overspeed a compressor head when output pressure is below 1/4 of max pressure? Is it cooling that's the issue? Because if it is, then I suspect it's a non-issue. The heat from compression at lower pressures will be less than at max pressure (right?) And in addition the cooling fan/pulley will be spinning so much faster than normal, that if anything, I suspect the compressor might actually experience cooler operating temps than normal.

But if the concern is oil starvation then I probably need to be daydreaming of something else entirely. I don't know if there's anything that can be done about that. There's probably also more important questions that I don't know enough to ask. So I petition this panel of online experts. What say you?

 

[FOMOGO]

I think you would see the same result as running your car for extended periods at 8k rpm, i.e. early demise. Compressors are designed to run at a fixed r.p.m., and the newer ones aren't built nearly as well as the older ones. Mike

 

[strantor]

I think you would see the same result as running your car for extended periods at 8k rpm, i.e. early demise. Compressors are designed to run at a fixed r.p.m., and the newer ones aren't built nearly as well as the older ones. Mike

Any insight into exactly why? Is it bearings? Is it something that could possibly be upgraded to withstand the new higher speed?

 

[silverhawk]

Air compressors work with a piston, and usually a single piston at that. There is no real way to balance that out so you can ramp up speed without occurring to much vibration. You will be replacing bearings often in the driving motors when you push the speed up, and likely even re- sleeving and changing pistons out with some regularity. If you stop to think about it, the technology is old enough that some one would be selling a high speed compressor for a lot of money by now if it could take the speed changes.

joe

 

[FOMOGO]

A combination of stroke, generally long, oil supply, often splash, and if pressure fed, generally only 10-15psi at speed, balancing of rotating assembly, minimal, ring pack, and piston to bore clearances, etc.. Mike

 

[Lo-Fi]

One key thing here: small compressors are usually splash lubricated, they do not have pressure lubricated big end bearings. The reciprocating forces climb steadily with rpm, which will overwhelm the oil film at some point. The connecting rod won't be designed for those forces either, and you may find that the reed valves won't like running that fast. A little way over rated rpm is going to be fine, but several times is going to end in disaster.

Look at it another way, though: on any given supply, running an induction motor is limited by the startup current. With the soft start the VFD offers, you can run a much larger compressor than you would on a switch. That's worth looking at!

 

[strantor]

Air compressors work with a piston, and usually a single piston at that. There is no real way to balance that out so you can ramp up speed without occurring to much vibration. You will be replacing bearings often in the driving motors when you push the speed up, and likely even re- sleeving and changing pistons out with some regularity. If you stop to think about it, the technology is old enough that some one would be selling a high speed compressor for a lot of money by now if it could take the speed changes.

joe

Most of the compressors in the HP range that I'm looking at are 2 cyl. And even the single cylinder models I believe have a counter weighted flywheel. Still, I'm sure being designed for such low RPMs, they are not balanced to the specs that would be called out for something spinning 4x as fast. Maybe the speed range could be improved by pulling the crank+flywheel and doing a very exact balancing of it?

You mention having to re-sleeve the compressor often. Why is that? Does higher speed cause more wear per stroke? Because this increased speed would not equate to more strokes in the lifetime of the compressor. Same number of strokes to reach full tank, we just get there faster.

The motor bearings can be upgraded. Whether or not the rotor will hold together in one piece is another story. Many vector duty motors these days are built to go well above 60hz and that's what I would be using. Pushing them to 240hz is probably a stretch but 120hz is quite common. I used 240hz (4x speed) as an example and I probably should have used a more realistic example.

Your point about old technology and someone would be doing it already, is a pretty safe point to make, but there are a lot things that go into it. The VFD and vector duty motor are expensive; things most consumers would probably not pay extra for. I just happen to have them laying around waiting for a purpose. And this would almost certainly have to be a consumer-level product. In the industrial market, there are screw compressors which would beat the pants off of any piston compressor in terms of reliability/longevity.

 

[ddillman]

go with a screw compressor it would hold up to the higher rpm's

 

[Ulma Doctor]

you would do better trying to figure how to get a larger volume pump, rather than trying to overspeed a small pump to gain performance.
if you overspeed a small pump, there is less time to draw in air to be compressed
the overheated air will transfer that heat into the compressor, to attempt to dissipate the best it can
the air charge will be less dense and require more volume to occupy the same space
the charge will most likely condensate and now you have a higher chance of water contamination and the stresses that causes

i can go on but, hopefully the picture has been painted

 

[strantor]

One key thing here: small compressors are usually splash lubricated, they do not have pressure lubricated big end bearings. The reciprocating forces climb steadily with rpm, which will overwhelm the oil film at some point. The connecting rod won't be designed for those forces either, and you may find that the reed valves won't like running that fast. A little way over rated rpm is going to be fine, but several times is going to end in disaster.

Look at it another way, though: on any given supply, running an induction motor is limited by the startup current. With the soft start the VFD offers, you can run a much larger compressor than you would on a switch. That's worth looking at!

Ok reed valves, that's a gotcha that I don't have an answer for. Also the inertia of the connecting rods, that's a stump. Thanks for the brain fodder.

Sooo.... let's see. I suppose I could oversize the compressor head in relation to the motor and gear it down. Since I already used 5hp and 240hz as my example I'll just continue with those numbers... instead of a 5hp motor and a 5hp compressor head, I use a 5hp motor and a 20hp compressor head. I use 4:1 double-reduction between motor and compressor. At start-up, the motor is spinning 4x its nameplate 60Hz RPM and the compressor is spinning exactly at its rated RPM. At start-up the 5hp compressor is delivering the volume of a 20hp compressor (like 60 CFM vs 12CFM). As pressure builds, the motor motor slows down as before, and as pressure approaches max, the compressor has slowed to where it's delivering about what what a 5HP compressor should be delivering right before the pressure switch cuts out.

How bout that?