Starting an air compressor SLOWLY?

Status
Not open for further replies.

moonshineJ

Member
Location
USA
Are there ways to start the aircompressor “slowly” to prevent voltage sag?

240 VAC, 3-phase Quincy 370 draws 210-220 A upon start up, which is normal if to look at the motor data plate. The problem is that it creates voltage sag, no matter if it’s fed from a generator,or from local electrical supply. Sometimes it becomes a nuisance, especially when voltage drop shuts some LVP controllers.

The local folks replaced the unloader,but it didn’t help a bit. And I don’t think that changing pulleys will help at all, either.

I’m wondering if anyone had to deal with similar problem. It’s a big load/inertia at startup, which needs to be overcome. Some VFD, or “soft start” controller? Any thoughts?
 

Attachments

  • Quincy 370.jpg
    Quincy 370.jpg
    146.1 KB · Views: 1

kingpb

Senior Member
Location
SE USA as far as you can go
Occupation
Engineer, Registered
Generator is to small if causing problems, unless cable is undersized.

Soft starter won't help if the starting torque is too high; which is one of the reasons with using them.
 

moonshineJ

Member
Location
USA
Generator is to small if causing problems, unless cable is undersized.

Soft starter won't help if the starting torque is too high; which is one of the reasons with using them.

Well, it's not just a generator. It's a small vessel; when they are on shore power, they experience the same problem.
 

moonshineJ

Member
Location
USA
VFD is an option. It's probably splash lubricated so you don't want it turning slow for very long.

Installing a bypass valve instead of a pressure switch so the motor just idles instead or stopping and starting is another trick.

Those guys replaced the unloader, thinking it can help, but it didn't. When they took belt off compressor, they say there is still voltage sag. From my understanding, it's not much different when they run compressor.

Having red all replies, it seems that cabling could be undersized? The gen is rated at 316 A, and startup current is 210-220 A.

I'm just trying to figure out all "pros" and "cons" of VFD for this application. I suggested folks to ask about VFD or "soft starting" the visiting tech who serves air compressors. He said that tipically it's not used for such applications.
 
Last edited:

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
Those guys replaced the unloader, thinking it can help, but it didn't. When they took belt off compressor, they say there is still voltage sag. From my understanding, it's not much different when they run compressor.

Having red all replies, it seems that cabling could be undersized? The gen is rated at 316 A, and startup current is 210-220 A.

I'm just trying to figure out all "pros" and "cons" of VFD for this application. I suggested folks to ask about VFD or "soft starting" the visiting tech who serves air compressors. He said that tipically it's not used for such applications.

A solid state soft starter IS the way to go if VD is the issue, most compressor suppliers have deals with soft starter mfrs to brand-label them or package them because it is such a popular option for air compressors.

That said, 15HP is usually too small to need it. So if you are having a significant voltage drop on a 15HP motor powered from a 360A generator even if the motor is UNCOUPLED from the compressor, there is some other serious problem. Insufficient conductor size jumps to the top of the list, meaning all of the conductors connecting to the starter and the motor. Your conductors for that entire circuit should be no smaller than #8, larger if the distance is longer.

In descending order of likelihood after that:

2) A problem with the motor starter, as in one contact is bad and although not completely single phasing the motor, it is not allowing it to develop full torque.

3) Your motor is 230/460V, is it possible that someone connected it wrong and wired it for 460V, but is only giving it 230V? Because if that happened, the motor would develop only about 25% peak torque and begin to stall, even unloaded, while still pulling high amps. If it WAS working fine before and nothing has been changed or removed, forget this possibility.

4) You have a winding going bad inside of the motor. It is drawing more current than the work it is doing represents.

5) You have a bad termination or splice somewhere.

Again, if there is a significant voltage drop when starting that little motor UNCOUPLED, then forget all of the mechanical issues and find the ELECTRICAL problem. A VFD or soft starter will likely make it worse by the way, it's something that needs to be fixed.
 

Bugman1400

Senior Member
Location
Charlotte, NC
I think Jraef has all good ideas. Other than that, I think your shore power source may have very high source impedance. If that is the case, bigger conductors won't make a difference. Perhaps you could use a voltage logger to see how bad the voltage dips when you start the compressor. If you have similar issues when on generation then you may need to get a better voltage regulator for your genset. If you don't have significant frequency sag when the compressor starts but, you do have voltage sag then I'd suspect the regulator.
 
Last edited:

Cow

Senior Member
Location
Eastern Oregon
Occupation
Electrician
I always use a couple different meters and check voltage at the service to rule out any possibility of meter error, then I go from there. If it dips hard at the service than a call to the power company is in order. There is a table Iwire has posted in the past that shows the normal range of utility voltages and the tolerances allowed.
 

kwired

Electron manager
Location
NE Nebraska
Does motor reach full speed and seem to have normal torque once it gets there?

Still may just be too high of source impedance (both sources) or too small of supply conductors for length or something of that nature that is only an issue when the higher startup current is present - this could include poor connections or contacts in the controller as has been mentioned as well.
 

moonshineJ

Member
Location
USA
I apologize that I reply to everybody at once. I think a few of you are right pointing toward cables' size.

I forgot to add from beginning, that they have two compressors, they both give the same symptoms. And they NEVER start at the same time, this would be too bad for sure. So, this would probably eliminate what Jraef ranked as #2 - a bad contact inside the starter, since both compressors create an issue.

I also eliminate #3.

#4, #5 - unlikely - again, since both motors behave the same.

Phil, during run compressors work just fine.

"A VFD or soft starter will likely make it worse by the way, it's something that needs to be fixed" - I absolutely agree on this one.

I tried to figure out how long this was like that. It seems forever, since compressors were installed in 1998. When I asked if they experienced LVP controller drops, like sewage pump, all this time I didn't get certain answer, but I'm pretty sure it could not be too bad all this time.
I'll tell them to get an electrician on site and check AWG, splices, contactors, and alike.
Shore installation may also have issues but you are right saying that the major culprit is electrical wiring onboard.

As of logging events, I thought about this, actually having on mind a different case. Problem is they have no electrician....unless something is going on. I found some good stuff on market, namely Hantek 365. It has models A thru F, with a price range $100-150. Going to get one.
 

kwired

Electron manager
Location
NE Nebraska
I apologize that I reply to everybody at once. I think a few of you are right pointing toward cables' size.

I forgot to add from beginning, that they have two compressors, they both give the same symptoms. And they NEVER start at the same time, this would be too bad for sure. So, this would probably eliminate what Jraef ranked as #2 - a bad contact inside the starter, since both compressors create an issue.

I also eliminate #3.

#4, #5 - unlikely - again, since both motors behave the same.

Phil, during run compressors work just fine.

"A VFD or soft starter will likely make it worse by the way, it's something that needs to be fixed" - I absolutely agree on this one.

I tried to figure out how long this was like that. It seems forever, since compressors were installed in 1998. When I asked if they experienced LVP controller drops, like sewage pump, all this time I didn't get certain answer, but I'm pretty sure it could not be too bad all this time.
I'll tell them to get an electrician on site and check AWG, splices, contactors, and alike.
Shore installation may also have issues but you are right saying that the major culprit is electrical wiring onboard.

As of logging events, I thought about this, actually having on mind a different case. Problem is they have no electrician....unless something is going on. I found some good stuff on market, namely Hantek 365. It has models A thru F, with a price range $100-150. Going to get one.
I don't agree the VFD or soft starter will make it worse. Controlled starting without having major impact on voltage to other loads is main reason for reduced voltage starting. That is about the only reason POCO require reduced voltage starting methods on larger motor installations.

I do agree if you have a problem with the motor it may only make things worse though.

You seem to have a source or situation with the supply conductors with limitations on what can be delivered when the motor is starting.

The generator is not as much of a surprise but if shore power can't get it done you must either have a long/undersized run and/or a undersized source to begin with.
 

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
moon -
First: kb hit it right on. Do not install a soft start or VFD until you KNOW you will have sufficient torque to start the compressor. As you said, 'It’s a big load/inertia at startup, which needs to be overcome."

The LRC you quoted meets spec for a Code J. The motor is performing as it should. Consider: Small conductors going to the motor will not increase the LRC. Installing larger conductors to the motor will not lower the LRC. Either way - doesn't matter, the conductor size is not the cause of the voltage sags that are dropping out the LVP controllers - unless the LVPs are out at the end of the motor feeder. (Q: What is LVP? Low Voltage Protected? As in drop out at a pre-set low voltage?)

The generator is not too small. We don't know the distribution voltage. 316A, .8pf,
208V --> 90KW
240V --> 105KW
480V --> 210KW
any of those is big enough to adequately start a 15hp motor.

Considering the evidence you have given us, the problem is a distribution issue. The motor LRC is bringing the problem to light - but it is not the villian. Forget the motor for a bit. It is just another load.

The original question was about preventing voltage sag at the LVP scontrollers. Lets take a look at the distribution path.
Is the generation at 480V , 240V, or 208V?
Is the shore power input at 480V, ... ?
What I'm looking for here is:
Is there a transformer between the source and the motor.
Is there a transformer between the source and the LVP loads?​
You will need to sketch out a one-line diagram. Start with the generator, to panelboards, to transformers, to loads.

You are looking for the place where the feed to motors splits off from the feed to the LVP starters - cause the voltage drop is upstream from there.

If I were guessing, I'd say the distribution was designed and in place long before 1998. The compressors were upgraded (bigger motors, biggers compressors) - but the distribution was not.

Get a one-line. Draw it your-self. That is the first step in troubleshooting exactly where the voltage drop is occuring.

Notes on the motor (mostly wild guesses):
The existing motor is an oddball - design A, code J. It doesn't have any better starting torque than a design B, but the LRC is undefined and often high - as you can see by the code J Locked Rotor KVA. Generally Design A is picked to trade off higher starting current for higher efficiency and low slip. (reference Nema MG-1). You might call the compressor mfg and ask them, "Why". But I doubt after 17 years anyone will know.

If I really had to drop the LRC, I'd consider new Design B, Code G motors. That will definitely drop the LRC and won't compromise the starting torque. DONOT get energy efficient - that just jacks up the LRC.

You might do it with a VFD. It can be programmed to hold up the starting torque. However, there are a new set of issues for designing a high starting torque vfd application:
It won't be done with a two-bit 15hp generic vfd. It will be one capable of 175% - 225% FLA for the starting duration.

Good luck - I'll be interested in what you find.

the Worm
 
Last edited:

moonshineJ

Member
Location
USA
Yes, LVP is the Low Voltage Protection.
The distribution voltage is 240 VAC, 3 phase from gens, and the drawing shows shore tie voltage as 230 VAC, 3 phase.
So, the gens make 240 VAC, 3 phase à 105 kW max. à 368 A max, 0.8 PF
No, there is no x-former between the incoming shore tie voltage and the motor, or between the generator bus and the motor.
From the drawing I see that all motors like fire pumps,A/C compressor, air compressors, water pumps, fans, and even kitchen equipment use 3 phase, 240 VAC. I’m sure they have some transformers for lighting and receptacles, etc. They mentioned a sewage pump; I don’t see it on drawing, but it may be 120 VAC, single phase.
Yep, the electrical distribution there is old, and the guy described it as “mess”. He’s not an electrician but I eagerly buy that.
Hopefully I go there at the end of month an find some skeletons in a closet.
 

moonshineJ

Member
Location
USA
Alright, here is what I saw so far:

- the shore tie cable (LSTHOF-150, AWG = 3/0, 3 conductors) is at least 3 times as long what "normal" shore tie would be (150')
- probably 35-40% of this length is coiled around metal rack where the shore tie receptacle is. Not sure how bad this sloppy inductance can be (well, some 9-10 loops, say 18' each) but when I saw those loops arounf steel rack I thought of an inductor at once.
- the shore tie breaker (200 A) is hot. Not that hot that my spit boils, but the side where UV coil should be appears white (116 F) in thermal imager. Going to inspect it tomorrow, but already told guys they need a new one.

I'm going to check resistance of all three conductors, but given 3/0 AWG, I'm sure I won't get something above 1 Ohm. Yet, starting current of each air compressor is some 210 A, so these useless 300' or more of tie + (perhaps) inductance can drop some chunk of the shore tie 240 VAC??

By the way, when air compressor started, I saw voltage sagged from 240 VAC to some 205 VAC.
 

kwired

Electron manager
Location
NE Nebraska
Alright, here is what I saw so far:

- the shore tie cable (LSTHOF-150, AWG = 3/0, 3 conductors) is at least 3 times as long what "normal" shore tie would be (150')
- probably 35-40% of this length is coiled around metal rack where the shore tie receptacle is. Not sure how bad this sloppy inductance can be (well, some 9-10 loops, say 18' each) but when I saw those loops arounf steel rack I thought of an inductor at once.
- the shore tie breaker (200 A) is hot. Not that hot that my spit boils, but the side where UV coil should be appears white (116 F) in thermal imager. Going to inspect it tomorrow, but already told guys they need a new one.

I'm going to check resistance of all three conductors, but given 3/0 AWG, I'm sure I won't get something above 1 Ohm. Yet, starting current of each air compressor is some 210 A, so these useless 300' or more of tie + (perhaps) inductance can drop some chunk of the shore tie 240 VAC??

By the way, when air compressor started, I saw voltage sagged from 240 VAC to some 205 VAC.

How much does voltage drop at the generator though when the air compressor is starting? Very likely you have some drop in the generator itself, then some more drop in the conductors between source and load. You can increase conductor size all you want, it will not change the impedance of the source.
 
Status
Not open for further replies.
Top