Dirty power - The lathes are knocking out my air conditioner

J&H · Jul 26, 2019

So we've got a Carrier Infinity AC unit in the shop. It's a full variable speed compressor system and it works pretty good. Whenever it's running at a higher compressor setting, it will set a high compressor current alarm whenever one of the lathes ends it's cycle and the spindle ramps down. It kicks the compressor out for about 10 minutes before it restarts at which time the next ending lathe cycle kicks it out again. After a few times of this, the unit locks out the compressor all together.

I know in the past, I've put a freq meter on the panel and noticed that the frequency really jumps up from 60hz when the VFDs are braking the spindles. I'm assuming this is causing the issues with the AC unit. I can clamp an amp meter on the line going to the AC unit and it is't using 8.5 amps at full load.. as one of the lathes is braking the spindle it will push the meter up to 10-10.5 amps for that couple of seconds.

So what do I need to put on the line to the AC unit to smooth that out? Who makes what kinda line conditioner to deal with this??

Ziggy2 · Jul 26, 2019

J&H said:
So we've got a Carrier Infinity AC unit in the shop. It's a full variable speed compressor system and it works pretty good. Whenever it's running at a higher compressor setting, it will set a high compressor current alarm whenever one of the lathes ends it's cycle and the spindle ramps down. It kicks the compressor out for about 10 minutes before it restarts at which time the next ending lathe cycle kicks it out again. After a few times of this, the unit locks out the compressor all together.

I know in the past, I've put a freq meter on the panel and noticed that the frequency really jumps up from 60hz when the VFDs are braking the spindles. I'm assuming this is causing the issues with the AC unit. I can clamp an amp meter on the line going to the AC unit and it is't using 8.5 amps at full load.. as one of the lathes is braking the spindle it will push the meter up to 10-10.5 amps for that couple of seconds.

So what do I need to put on the line to the AC unit to smooth that out? Who makes what kinda line conditioner to deal with this??

Your lathes are using dynamic braking to slow the spindle speed on the ramp down. The excess energy needs to go somewhere so the lathe VFD puts the excess onto the incoming AC line via increased voltage.

You could try increasing the deceleration time which will lower the voltage increase that the HVAC sees.

Your air conditioner should have a VFD for the blower and a two speed compressor. Running the compressor in low speed would also help.

Your problem really isn't dirty power, which a sine filter could help. Now if you had a high inertia load that was running all of the time, i.e. an idler motor with a flywheel running off the mains, that could absorb the decell energy and control the voltage increase, that would probably solve the problem.

J&H · Jul 26, 2019

Ziggy2 said:
Your lathes are using dynamic braking to slow the spindle speed on the ramp down. The excess energy needs to go somewhere so the lathe VFD puts the excess onto the incoming AC line via increased voltage.

You could try increasing the deceleration time which will lower the voltage increase that the HVAC sees.

Your air conditioner should have a VFD for the blower and a two speed compressor. Running the compressor in low speed would also help.

Your problem really isn't dirty power, which a sine filter could help. Now if you had a high inertia load that was running all of the time, i.e. an idler motor with a flywheel running off the mains, that could absorb the decell energy and control the voltage increase, that would probably solve the problem.

Thanks for the reply..

We did increase the deceleration time on one of the lathes, but it didn't seem to make much difference. It looks like we would have to make a significant change in deceleration time to make an impacting difference..

The compressor in this thing is 100% variable, not just two speed. I've tried to limit the compressor to say 50%, but there's no settings in the control to make that happen and carrier says there is no way. You can limit the max compressor RPM, but it really only limits it in heat pump mode. You can't select an RPM that's lower than 100% of it's cooling speed.

How big of a motor are we talking to absorb the power? 5-10HP? That could be something I could try without too much trouble.

Edit: I have put a DMM on the line and I don't really see an increase in the voltage to the AC unit... But perhaps it happens faster then the meter can react?

Edit edit: We also wired through a buck/boost transformer we had lying around and cut the voltage down from 245 to 223 and it made no difference.

Ziggy2 · Jul 26, 2019

J&H said:
Thanks for the reply..

We did increase the deceleration time on one of the lathes, but it didn't seem to make much difference. It looks like we would have to make a significant change in deceleration time to make an impacting difference..

The compressor in this thing is 100% variable, not just two speed. I've tried to limit the compressor to say 50%, but there's no settings in the control to make that happen and carrier says there is no way. You can limit the max compressor RPM, but it really only limits it in heat pump mode. You can't select an RPM that's lower than 100% of it's cooling speed.

How big of a motor are we talking to absorb the power? 5-10HP? That could be something I could try without too much trouble.

Edit: I have put a DMM on the line and I don't really see an increase in the voltage to the AC unit... But perhaps it happens faster then the meter can react?

Edit edit: We also wired through a buck/boost transformer we had lying around and cut the voltage down from 245 to 223 and it made no difference.

Yes, a regular DMM will not show the voltage transients you are experiencing.

Considering what you have tried up to this point and the fact that your HVAC is actually a variable speed compressor, your best solution will probably be a sine wave filter in front of the HVAC.

My initial advice was with the assumption that you had only a two speed compressor. All of the consumer fancy stuff does not always like to play nice in an industrial environment.

deltap · Jul 26, 2019

Do you have an AC line reactor on the vfds?

J&H · Jul 26, 2019

deltap said:
Do you have an AC line reactor on the vfds?

I've not added anything to the lathes. Both are Mazaks. A QT-15 and a QTN-200-II.

n2zon · Jul 26, 2019

J&H said:
So we've got a Carrier Infinity AC unit in the shop. It's a full variable speed compressor system and it works pretty good. Whenever it's running at a higher compressor setting, it will set a high compressor current alarm whenever one of the lathes ends it's cycle and the spindle ramps down. It kicks the compressor out for about 10 minutes before it restarts at which time the next ending lathe cycle kicks it out again. After a few times of this, the unit locks out the compressor all together.

I know in the past, I've put a freq meter on the panel and noticed that the frequency really jumps up from 60hz when the VFDs are braking the spindles. I'm assuming this is causing the issues with the AC unit. I can clamp an amp meter on the line going to the AC unit and it is't using 8.5 amps at full load.. as one of the lathes is braking the spindle it will push the meter up to 10-10.5 amps for that couple of seconds.

So what do I need to put on the line to the AC unit to smooth that out? Who makes what kinda line conditioner to deal with this??

Put a big (brick size) MOV on the AC circuit breaker?

DDoug · Jul 26, 2019

The lathes have become sentient, and are taking control... :skep:

They don't need A.C., they want more power, and there are qty (2) of them and qty (1) of you..... :toetap:

Bill D · Jul 26, 2019

Do the lathes have braking resistors to dump the energy into at slowdown? I would suggest a line reactor on the ac unit. If that does not work a load reactor on both lathes.
Bill D

I have seen cheap no name VFD's with braking resistor terminal screws that are not electrically connected to anything inside the VFD. Makes owners warm and fuzzy to add a braking resistor but it does absolutely nothing.

thermite · Jul 26, 2019

if you have ONE problematic AC and more than one expensive machine tool, I'd peddle the fancy Carrier for whatever I could get for it, used, and replace it with a DUMB single-speed AC unit that simply didn't give a flying f***k what the lathes were up to.

York. Old-skewl Brisol recip, not even "scroll" compressor in my case,

MY 4Q, FULLY regenerative Dee Cee Drives make even the nastiest of VFD's look Snow-White virtuous by 'scope-recorded comparison, dirty buggers as they can be.

ANYTHING else is just going to cost more money and take more time messing about.

Honour the threat. Smite the problem, not the bystanders.

It JF works better that way, even if it is not Ingineering-"Elegant" nor "politically correct".

PS: Competent HVAC guru can simply swap-out the guts & controls. "downgrade" to higher reliability, same as if repairing an over-age-in-grade system. Done all the time.

Tony Quiring · Jul 26, 2019

Possible the driven motor being braked not running at an rpm such to generate 60 hz in braking, no clue how yours works.

Could be generating out of sync energy that compressor vfd is rejecting.

Line reactor or other conditioner on the a.c. would be next.

Possible consult to a.c. manufacturer support.

Sent from my SAMSUNG-SM-G930A using Tapatalk

rogertoolmaker · Jul 26, 2019

Sound like you may have a poor ground.

johansen · Jul 26, 2019

its my hypothesis that the extra 2-3 amps that show up on the hvac unit are entirely ripple current due to the poor filtering on the input side of the vfd.
there is a good chance your hvac inverter unit is a power factor corrected front end, which means it has capacitors line to line at the front end, probably on the order of 10-20uf (in 2 or 3 stages with common mode chokes), and that's enough to make an extra few amps show up due to the high frequency ripple produced by the lathe's regeneration circuitry.

if the hvac inverter is a dumb rectifier type, then you would have to have enough emi to cause the voltage at the capacitors to rise substantially higher than 240vac/340vfd before it shuts down due to high voltage or some other problem manifests. i suspect it would run fine as high as 280vac.
But it could still be a dumb rectifier but with enough filter capacitance to cause your amp meter to read the ripple current amps.. in which case, its not the voltage that broke something but rather the electrical noise. i don't think its the rms ripple current from the lathe that is causing the hvac to fault but rather the common mode noise (much higher frequencies than 5khz in such a case.)

you may be able to fix the problem by simply adding 10-100uf line to neutral at the lathe's vfd supply, on all 3 phases.

if that doesn't cut it then add a line reactor at the hvac unit. capacitors line to line and line to neutral at the hvac unit, line reactor upstream of those caps.

no need to buy a line reactor to determine if you need one. a 100 foot roll of 12/2 romex with one of the wires connected backwards* should give you enough inductance to fix this problem.
*unless you run all three wires in series and put the inductor on only one side of the line. (otherwise the current cancels out)

its still possible that your hvac unit is a "dumb" rectifier and the extra amps you are measuring are entirely due to the voltage ripple caused by the lathe regeneration system (in which case, the average amps which charge up the capacitors are still the same, the hvac unit is in no danger of being destroyed) but, the hvac system is faulting because it has circuitry which measures the input line amps at the input of the rectifier. such circuitry is almost certainly not going to be rms reading but rather just a current transformer, diode and resistor, and the hvac system throws a fault because it thinks something has gone wrong.

if this is the case then simply adding 10-100uf line to line at the hvac system will probably fix the problem

thermite · Jul 26, 2019

rogertoolmaker said:
Sound like you may have a poor ground.

Could was, yes. Rendering the problem worse AND the Carrier's brains overly sensitive.

Unlike 4Q thyristor-class DC Drives that have no passive rectifier into a capacitor bank (basically a "one way" system), but DO have the extra controlled rectifiers to pass surplus power generated by braking back to the upstream, the "average" VFD is not "regenerative", anyway.

If the VFD's filter capacitor bank can absorb the over-run, fine.

Otherwise it has to be switched to a braking resistor. It takes significant extra circuitry - read MONEY - to build a VFD able to pump a surplus back into an AC grid AS AC, phase-matched and all, not just into a simple factory or zone wide DC bus shared amongst several "local" VFD.

True regen on a VFD is not generally needed, as the AC motors being driven as loads are not as effective as DC motors when acting as generators, anyway.

J&H · Jul 27, 2019

johansen said:
its my hypothesis that the extra 2-3 amps that show up on the hvac unit are entirely ripple current due to the poor filtering on the input side of the vfd.
there is a good chance your hvac inverter unit is a power factor corrected front end, which means it has capacitors line to line at the front end, probably on the order of 10-20uf (in 2 or 3 stages with common mode chokes), and that's enough to make an extra few amps show up due to the high frequency ripple produced by the lathe's regeneration circuitry.

if the hvac inverter is a dumb rectifier type, then you would have to have enough emi to cause the voltage at the capacitors to rise substantially higher than 240vac/340vfd before it shuts down due to high voltage or some other problem manifests. i suspect it would run fine as high as 280vac.
But it could still be a dumb rectifier but with enough filter capacitance to cause your amp meter to read the ripple current amps.. in which case, its not the voltage that broke something but rather the electrical noise. i don't think its the rms ripple current from the lathe that is causing the hvac to fault but rather the common mode noise (much higher frequencies than 5khz in such a case.)

you may be able to fix the problem by simply adding 10-100uf line to neutral at the lathe's vfd supply, on all 3 phases.

if that doesn't cut it then add a line reactor at the hvac unit. capacitors line to line and line to neutral at the hvac unit, line reactor upstream of those caps.

no need to buy a line reactor to determine if you need one. a 100 foot roll of 12/2 romex with one of the wires connected backwards* should give you enough inductance to fix this problem.
*unless you run all three wires in series and put the inductor on only one side of the line. (otherwise the current cancels out)

its still possible that your hvac unit is a "dumb" rectifier and the extra amps you are measuring are entirely due to the voltage ripple caused by the lathe regeneration system (in which case, the average amps which charge up the capacitors are still the same, the hvac unit is in no danger of being destroyed) but, the hvac system is faulting because it has circuitry which measures the input line amps at the input of the rectifier. such circuitry is almost certainly not going to be rms reading but rather just a current transformer, diode and resistor, and the hvac system throws a fault because it thinks something has gone wrong.

if this is the case then simply adding 10-100uf line to line at the hvac system will probably fix the problem

I feel like this is all an accurate description of what I'm seeing. When you say add 10-100uf to the hvac system... are you meaning a single capacitor in the 10-100 range or ten 100uf caps together? I'm assuming a single cap within that range... I can probably try that today. Also, it seems like a line reactor isn't very much money to try.

Scottl · Jul 27, 2019

See if you can find some split ferrites to clamp around the power wires both at the source of the noise and the compressor wires. What these do is act like a shorted turn at high frequencies and attenuate some of the EMI peaks.

Back in the analog TV days I had a dishwasher that would put stripes of noise on the screen at several portions of the cycle. I put split ferrites around the ac leads and at a couple of solenoids - interference gone. you can also use non-split ferrite cores if the wires can be easily disconnected. This is standard practice in industry to make products pass emissions/immunity tests and you've probably seen PC monitor cables with a lump near one connector. That's a ferrite core with over-molding to make it look nice.

thermite · Jul 27, 2019

J&H said:
I feel like this is all an accurate description of what I'm seeing. When you say add 10-100uf to the hvac system... are you meaning a single capacitor in the 10-100 range or ten 100uf caps together? I'm assuming a single cap within that range... I can probably try that today. Also, it seems like a line reactor isn't very much money to try.

Your electrican can install the least-costly attempt at it right inside the load center. Stock items, all major makers, for taming near-miss thunderstorms, up-grid. About 50 bucks. Good to have ANYWAY.

Then prices gradually climb with aggressiveness and effectiveness.

My favourit here, is an EGS Hevi-Duty 27 KVA "drive isolation" transformer!

dazz · Jul 27, 2019

Hi

You need to "see" what is happening on the supply. Find someone with a digital oscilloscope who can simultaneously measure current and voltage on the supply line. This is not a piece of gear your average electrician will have, or know how to operate on high voltage equipment, or who can interpret the data.

There is a good chance the HVAC logs events for diagnostics. Ask the Carrier guy if they can access any fault logs. They should tell you why the HVAC is tripping.

If the lathes are sending volt spikes back into the supply, there is a good chance the lathe VFDs are being over-stressed. If so, that might shorten the life of the VFDs.

Until you can get some good measurements with the right test gear and the right operator, you will be groping in the dark.

Dazz

J&H · Jul 30, 2019

So we've tried a few caps line to line with no difference. I got ahold of a Sola HD 5000va line conditioner and have it feeding the AC unit and it's working now, the lathes don't bother it one bit anymore.

The downsides are that the conditioner is fairly loud, not a big deal when the machines are running though and it also pulls about 2 amps continuous. That kinda kills the efficiency of the fancy AC unit. I may wire in a couple switches so it can be bypassed when the lathes aren't running or won't run for a couple days...

Tony Quiring · Jul 30, 2019

Some he contractors can be wired to do just that.

24 VAC coils that are connected to thermostat can connect ac only when call to cool.

Another set can connect conditioner whenever lathe is running.

To bypass conditioner a reversing motor starter would work as it is a pair of contractors that are interlocked so only one activates at a time.

Sent from my SAMSUNG-SM-G930A using Tapatalk

Dirty power - The lathes are knocking out my air conditioner

Aluminum

Stainless

Aluminum

Stainless

Cast Iron

Aluminum

Cast Iron

Diamond

Diamond

Diamond

Titanium

Cast Iron

Stainless

Diamond

Aluminum

Diamond

Diamond

Stainless

Aluminum

Titanium

Similar threads