Reducing start up current on air compressor for off grid

oil pan 4

Trying to inch my way off grid and at least save some power in the mean time I have 240v air compressors that have starting inrush current of at least 45 amps according to my laggy amp meter. It may be higher than 45 amps but that is what my meter is showing so that is what I will use as my standard for measurement and its not tripping a 30 amp breaker. This compressor normally shuts off at a little under 20 amps.
So I am finding that just normally starting these air compressors they are hitting 40 to 45 amps. That seems like a lot of amps for an off grid power inverter.
So to reduce starting inrush current I am looking into building some kind of soft starter.
For testing I got my 0-300v 20amp 5kw varrac wired up with new cord, plug and switch.
After some testing I found that soft starting at about 220v (normal voltage here is usually right about 250v) reduces starting amps from 45 down to 35 amps saving 10 amps. No soft starting trickery I tried could do any less than 35 amps, which is still a lot.

Next I tried across the line starting a 240v motor with out a belt on it, that only peaked at 20 amps, which is a lot better than 45 amps. To try and further reduce amps from that I put it on my varrac and started the motor at lower and lower voltages. Eventually I found free spin starting the motor at 150v reduced the startup to 10 amps. But if I am going to regularly use starting the motor unloaded, in free spin I have to find a way to engage and disengage the belt automatically.
Tomorrow I am going to try free spin starting the motor then engage the belt with the motor running and see how much the amp draw goes up while spinning up the compressor. I have a feeling it will stay way under 45 amps, even well under 35 amps, I think it will even stay under 20 amps. It may not go much over 10 amps.
So I'm thinking I may be able to reduce start up current down to about what the running amps are. The only way I can do this with single phase is before start up have the belt slacked, start the motor on a closed circuit resistive soft starter on a time delay, tension the belt and run up the air compressor to the set pressure, turn off and slack the belt, all automatically.

I could fire up my 5.5hp gasoline air compressor or fire up my 17.5kw generator but that kind of a lot of gas and makes a lot of noise. Or I could get a 3 phase motor and VFD, but that's not till later on, I will call that plan D, for VFD.

mike95490

The quest for the holy grail.
It's called a 3 phase motor, and a compressor unloader (air valve that lets the compressor spin up with no backpressure)
Then you get a VFD (variable frequency drive)  controller and let your single phase 240V power that,  Grainger carrys them

mcgivor

The question I would ask first is, what is the compressor used for, if it is air tools, then that would be about the most inefficient use of energy possible, running off grid would need an array that is capable of supplying the energy required to run the compressor and a battery to cover the start up  requirements, using it only while the sun shines. Each HP would need about 707W of array, ignoring efficiency losses, and an inverter large enough to absorb the inrush. Currently I have a 1 HP 230V well pump which I have used with a 2000W inverter ( 4000w surge )  and a 1500W array which works for a while when batteries are fully charged, off the array, but tends to end up taking from the battery after a while, so I now run a generator to pump water. My guess is it would take about 2000W of array to cover each HP of motor demand, in full sun conditions.

As far as reducing inrush short of the 3 phase VFD route, unloading the compressor head as Mike has mentioned is one way, a centrifugal clutch could be another, or both in combination. Attached link explaining air tool inefficiently, may or may not  be of interest. 
http://fluidpowerjournal.com/2014/04/air-tool-is-inefficient/

oil pan 4

I know all about VFDs. Eventually I'm sure I will get at least one air compressor on a 3 phase motor and VFD with unloader tied into a shutdown timer.

The compressors is used for everything from sand blasting, painting, air for a plasma cutter and some air tools but they don't use a lot of air. Almost everything that can be made electric already is. There is no electric replacement for sand blasters, painters (electric painters suck) or air for a plasma cutter.

Powering from solar may be unworkable so I may end up powering them with my 17.5kw or 7kw generator or use my 5.5hp gasoline air compressor. Then the difference is being able to power 1 air compressor with my 7kw with simple high inrush starting or be able to run 2 air compressors with soft start on the 7kw generator as opposed to dragging out the 17.5kw to run both.
Or being able to run the plasma cutter and the air compressor on 1 generator, I have not tried it yet but across the line starting my air compressor where it draws 45 while the plasma cutter is going may cause the plasma cutter to fault for input voltage, even on my 17.5kw generator.

dennis461

Make sure the air compressor un-loading valve is working properly.  You have one, right? You may also be able to increase the size of hose/line between the compressor and the tank which should help starting ..if the tank has an inlet check valve.

oil pan 4

Yes all that stuff works.
I built these compressors my self. They have a check valve, unloader and they have about 3 feet of 5/8 line between the compressor discharge and the tank inlet check valve. I even spent money on new safety pressure relief valves.
I have considered adding a high volume unloader solenoid valve on a time delay to keep pressure from building up, but I need to test this first to see if it even helps. To test it I would just unhook the discharge line and start it up and see if it makes a difference.

oil pan 4

I have also power factor corrected my air compressors, it only saves 2 to 3 amps, but off grid that is kind of huge.

dennis461

Soft starters are available off the shelf.  I do not however have experience with soft starters coupled with inverter input.
EasyStart™ 364 (3-ton)

Schneider Electric ATS01N112FT

excuse the font's

oil pan 4

I have seen the single phase A/C unit soft starters. They do not appear to work with capacitor start capacitor run motors.
Soft starting the air compressor's capacitor start capacitor run motors while the compressor drive belt is tensioned with my varrac I was only able to reduce starting amps from 45 down to 35.
I think the motor, pulleys and the compressor is just too much inertia for a single phase motor to soft start.

oil pan 4

Since I know I can free spin start a single phase motor with as little as 10 amps on my varrac I just need to figure out what the resistance is and use resistors because I am not dedicating my varrac to be a motor starter I'm thinking I should use wire wound resistors, hot water heater elements or clothes dryer heating elements.
I figure when across the line free spin starting starting the motor where it draws about 20 amps it has about 12 ohms of resistance. Free spin soft starting where the motor only draws about 10 amps it's more like having 25 ohms of overall resistance. Just need the resistors to drop the volts from 250 down to about 150, adding about a 12 ohm resistor should do the trick.

I have some wire wound resistors and hot water heater elements. A trip to the scrap yard should net me a dryer heating element or 2.
Dryer heating elements should have a resistance of around 10 ohms and so should hot water heating elements. My wound resistors I have would only add about 3 or 4 ohms if I series them up.

oil pan 4

I tested my resistive starter on a free spinning motor. It peaked at 16 amps. But the ramp up to 16 amps was nice and smooth.
Compared to across the line starting where it just instantly shoots up to between 20 and 22 amps.
My resistor is a dryer heating element. It provides the 10 ohms I was looking for and won't over heat.
That could work but I think can do better with inductive current limiting starting.

BB.

You could also (possibly) use an NTC (negative temperature coefficient) resistor: Higher resistance when cold, Lower resistance when hot.

http://www.newark.com/pdfs/techarticles/ametherm/Ametherm_PPT_FEB07_REV6DistyTutorial.pdf

-Bill

mike95490

BB. said:

You could also (possibly) use an NTC (negative temperature coefficient) resistor: Higher resistance when cold, Lower resistance when hot.
http://www.newark.com/pdfs/techarticles/ametherm/Ametherm_PPT_FEB07_REV6DistyTutorial.pdf
-Bill

That's what Iota uses in their power supplies, to tame the initial inrush, NTC's

oil pan 4

That could definitely work. I would just need to pick one that is the correct mass to heat up while starting.
This morning in think I will start off by gutting a microwave oven transformer, specifically removing the high voltage winding to make it less lethal.
There are 2 types of reactive starters. Air core and iron core. Conflicting sources describe both as being superior and more common than the other. So I guess that means both work.

This is what I know so far:
The compressor and motor belted together just have too much inertia to soft start with a single phase capacitor start capacitor run motor. It will burn up before it actually soft starts while tring to drive the compressor. I'm working against the very nature of these motors by trying to soft start them.
They are designed to go full torque on startup draw as many amps as needed or die trying.
I still think I am getting much higher amps then what I am seeing on my instrumentation when across the line starting with a load on the motor.

If money were no object space vector modulation on a 3 phase motor is definitely the way to go.

Dave Angelini

If money were no object I would use something like Mikes old XW6048

oil pan 4

I tried 2 different micro wave oven transformers that were almost the same size. I removed the secondary on both. One started kind of slow and peaked at 22 amps. The other started faster and peaked at 17 amps.
Auto transformers are best.

mike95490

Dave Angelini said:

If money were no object I would use something like Mikes old XW6048

No way !  I'm still using it.  I got it specifically for it's surge to start the well pump and fridges, without dimming lights in the house! I just hope when I have to replace it, the new gear works as well

Dave Angelini

Mike, the XW+ is actually better so no worries!

oil pan 4

I tried both reactors in series and the motor wouldn't even turn.
Next I will try them in parallel. I have one more I might decore that is a little bigger. A made in USA copper core from the 1980s.

oil pan 4

The most popular NTC resistor used for motor starting is graphite.
The most popular resistance starters just use metal wound wire.
Auto transformers are the most efficient reduced voltage starters per amp.

oil pan 4

I think I need to pick one.
I will just stick with what power companies like large industrial to use, since I am trying to be my own power company at some point. Obviously they like VFD but I'm not there yet.
Power companies don't like across the line starting because of the power fluctuations it causes.
Then it breaks down to open transition and closed transition
Power companies don't like open transition so much. Examples of open transition starters are wye-delta and auto transformer, with open transition the motor is briefly disconnected from the power source to switch from the soft start to full power. This switching causes transient power surges. I'm sure power inverters don't like violent current inrush or high voltage transients, but I'm not sure which is worse for them.
Then there is closed transition . That's where you have a resistor starter or inductive soft start device then just use a contactor to by pass the current limit device for full motor power. The circuit stays closed during the transition. Reactor soft start may cause some transients but those are easy to deal with.
So that pretty much narrows it down to resistors or reactors.
So a little more testing, waiting on some wire wound resistors to come in and I need to find some big load dump resistors I was using for my wind power experiments a few years ago.
And gut another microwave oven transformer or 2.
Then figure out how I am going to slack the belt and control it all. But one step at a time.

oil pan 4

Most of the microwave oven transformers I test have 0.4 to 0.7 ohms of resistance. I'm pretty sure that these are aluminum wire core.
My big 1980s transformer with a copper core has the expected 0.2 ohms you typically see with a copper core like this.

dennis461

I have a gasoline powered air compressor for sale cheap 
Have you thought of using a centrifugal clutch on the motor?

If you were starting from scratch, how would you build the system, maybe it's time to dump the single large induction motor piston compressor, go with two small screw compressors on a common header?

Could you use a timer to send 120VAC to motor for one-two seconds, then hit it with 240VAC?
Your electrical system is 240 split phase, right?

oil pan 4

I have a 5.5hp gasoline air compressor it makes around 15 to 17cfm at 90psi.
My 240v electric air compressors run at about 2.5hp and 3.5hp. They are kind if small.
I'm trying to make my existing equipment more off grid friendly.
Half voltage starting still draws a ton of amps. This type of starter would be an open transition starter, the open transition thing is bad.
I may try testing a centrifugal clutch, all I know is I need something to slack the belt or uncouple motor and compressor when starting the motor. Everything is kind of riding on being able to do that.
We use screw compressors where I work and they are pretty cool, only problem is they are very expensive.
I currently have split phase 240v going to the house.
Generators are center tapped neutral ground bonded 240v single phase. Current stand alone inverters are 120v, eventually I may end up getting a 240v split phase inverter, but that won't happen for a while.

oil pan 4

I tried to run my larger air compressor off my 7kw generator and it couldn't get it up to speed. The speed dropped down to less than 47hz. Then it tripped the thermal over load.
The 17.5kw generator didn't even notice when I plugged the compressor into it.

oil pan 4

I had an idea that 4 pole motors may use a little less start up current.
I have found that the 2 pole motors fairly consistently hit a momentary inrush around 200% of rated full load amps while full power free spin starting and 400% of FLA loaded.
The instantaneous inrush is much higher, maybe even 10x FLA.
4 pole motors are hitting around 130% to 150% on full power free spin startup.
And they are hitting under 200% FLA while doing a full power start while loaded.
That's not a little less that's a lot less power.

Unless you need the direct drive high-speed offered by a 2 pole motor, well they kind of suck. 4 pole motors are a far better choice well pretty much all around.

I was able to get a free 1hp 4 pole 3 phase motor. It appears to have good bearings and it checks good on a multi meter. I am going to do a quick and dirty hass-kamp conversion to make sure that it does work. Once I confirmed it works I'm going to order an ABB 240v single phase VFD to power it.

mcgivor

Was changing a belt on an automatic (CVT transmission) scooter earlier today, that could work in your application, but finding one ? perhaps a snowmobile. 

oil pan 4

I plan on getting a belt clutch for my gasoline powered air compressor to make cold starting a lot easier.
Before input the clutch on a gasoline compressor I may try putting it on an electric just to see what happens.

oil pan 4

I ordered a belt clutch for a go-cart. Only problem is if I want to use the cart clutch on a 4 pole motor the 3.2 inch diameter which is small for an 1800rpm motor.
I'm going to stick the cart clutch on a 2 pole motor powered compressor first to see if it helps much.
If it does work I will modify the clutch for 4 pole use. Alter the clutch springs to engage at a lower speed if need be and weld on a more appropriate 5 or 6 inch pulley.

Then I searched for a magnetic clutch for 5/8 shaft. A fruitless search, until now. I found a clutch that bolts on to a 56TC frame motor. So I bought it. Only problem is that I don't have a 4 pole 56TC frame motor but I can get one.
I tested slam engaging the 2 pole motor running at full speed to a stopped compressor the current surged up to about 200% of FLA or about 20 amps which is also about what full power free spin starting the motor peaked at.
I'm thinking slam starting the compressor on a 4 pole with a magnetic clutch wont draw as much power as a 2 pole. I think magnetic clutching a 4 pole to the compressor will draw more like 100 to 150% of FLA or roughly equal to full power free spin starting amps.

oil pan 4

Another addition, I picked up a 3 phase 1.5hp 56c frame motor, aka a gear case motor, it's got a bolt on pump face but no mounting base. It was free.
It's going to be powered by an ABB single phase input VFD, which is not free.

oil pan 4

I got the ABB drive and 3 phase motor working together. The drive is only working in manual at the moment.
This drive does have DC inputs so I should be able to start and stop the motor, change speeds using only pressure switches.

This drive also has a 4-20ma analog input and I found a 4-20ma pressure transducer that is scaled for 0 to 140psi. I may be able to run it at fully variable speed if I can figure out how to set it up the program on the drive.