Any electric engineers?

R

Racerx

Final Approach
PoA Supporter
Joined
May 15, 2020
Messages
7,764
Display Name
Ernie
Home Airfield
C81
Been having an ongoing issue for years at this point with my power at work. 240 delta 3ph. The condescending electrical engineer from the local power company blamed it on my equipment being to sensitive to higher voltage and promptly left.

Was about to cut something on the CNC plasma but the power source is giving an error. An error that means either the board is trashed or that I've got an input voltage issue. I check my crane that's on a step up transformer and sure enough, it makes a rattle going up and won't move north/south

The field supervisor or whatever his title is says whenever I'm switched over (whatever that means) that's when I have trouble. When it's normal everything works.

I have both a cheap multimeter and a better true RMS meter. The RMS meter says everything is good. And that's what the power company is saying that everything is good.

Line to line with the RMS meter I'll get 253-250-248.
Line to line with cheap meter I'll get 232-258-256.

When I'm switched back to normal the cheap meter and the RMS meter will match and everything runs happily. The EE's claim that my machines don't like high power is mularkey as previously I measured power 253-254-252 with both meters and everything ran fine. It's only when the meters don't agree and the cheap meter shows imbalance that I have issues.

Any thoughts to take to the power company or utility board?
 
I am only a nuclear engineer, but I'll toss my two cents in. Your power problem is likely more related to the un-balanced feed. I don't know if your two measurements followed the same phase relationship or not, but it appears one of the phases is too low. Three phase industrials generally(but not always) are indifferent to slight or even moderate voltage differences in the delta Y but some equipment is sensitive enough to be upset. A CNC machine is a good example of the type of machine that may not like a phase being 5% or 10% sagged.

Since you are fighting with a utility, or power supplier, you aren't going to win the battle until you prove they are out of spec. So - first you need to know what spec they are giving you, and what min, max and tolerance plus what distortions limits are. Also, resting voltages are not always representative of what's happening under loads, and switching loads.

What you need, and what's going to happen if you want to 'win' this argument is that you will rent something called a "Dranetz meter".

Dranetz Technologies - GlobalTestSupply.com

www.globaltestsupply.com www.globaltestsupply.com

Get the kind that has a recording tape or a graph storage of the output of the supply. Have it connected by a pro at the breakout where the power delivery changes to your equipment so you can say with accuracy that the graph shows the actual power, fluctuations, distortion, sag, spikes, etc that is coming from the power co. The only evidence they will accept is from this type of recording. Then, you can find out if you are getting what they promise, or if they are out of spec.

Compare this to the spec required by your CNC. Lot of work, but you can talk to the power co until you are at screaming stage, and they won't do anything without the proof.
Edit to add: Do NOT tell the power provider you are putting in a Dranetz monitoring, they may adjust their feed if they know you are recording. I've done this successfully at several power suppliers when I could show 40-60mS spikes of 298V on a max 240V leg.

Edit again; here's a rental place for the Dranetz:
www.atecorp.com

Dranetz HDPQ Xplorer Plus Power Quality Analyzer Rentals

The Dranetz HDPQ Xplorer Plus is an exceptional power quality analyzer with superior PQ monitoring abilities in a portable and durable package. The...
www.atecorp.com www.atecorp.com
 
Last edited:
Question 1 - switched over to what?

252 is normal, what's not normal is a phase being 10% less.
A multimeter can only tell you so much, what you need is a power quality meter, that can display the waveform and measure THD (total harmonic distortion).

I suspect you might have some loads on one phase only that have a bad power factor (current leading or lagging the voltage by a significant amount), and possibly high distorsions in the process.
 
AV8R_87 said:
Question 1 - switched over to what?

252 is normal, what's not normal is a phase being 10% less.
A multimeter can only tell you so much, what you need is a power quality meter, that can display the waveform and measure THD (total harmonic distortion).

I suspect you might have some loads on one phase only that have a bad power factor (current leading or lagging the voltage by a significant amount), and possibly high distorsions in the process.
Click to expand...
Per comed "The feeder is switched abnormally as of this morning. It's not scheduled to go back to normal until for a couple days".

When I test my power I shut the main off prior to testing to rule out any loads on one phase.

My problem is my RMS multimeter will read balanced and within tariff. However my equipment doesnt work and thats when I try my non rms meter which gives drastically different and imbalanced results.

Meter (A) 229-252-250
Meter (B) 252-244-247

The power company simply looks at the RMS meter readings calls everything good and blames my equipment and the cheap meter....But my equipment is agreeing with the cheap meter. This only happens when they switch feeders. When i'm on my normal feed the rms and non rms meters read identical.
 
Barrett50 said:
I am only a nuclear engineer, but I'll toss my two cents in. Your power problem is likely more related to the un-balanced feed. I don't know if your two measurements followed the same phase relationship or not, but it appears one of the phases is too low. Three phase industrials generally(but not always) are indifferent to slight or even moderate voltage differences in the delta Y but some equipment is sensitive enough to be upset. A CNC machine is a good example of the type of machine that may not like a phase being 5% or 10% sagged.

Since you are fighting with a utility, or power supplier, you aren't going to win the battle until you prove they are out of spec. So - first you need to know what spec they are giving you, and what min, max and tolerance plus what distortions limits are. Also, resting voltages are not always representative of what's happening under loads, and switching loads.

What you need, and what's going to happen if you want to 'win' this argument is that you will rent something called a "Dranetz meter".

Dranetz Technologies - GlobalTestSupply.com

www.globaltestsupply.com www.globaltestsupply.com

Get the kind that has a recording tape or a graph storage of the output of the supply. Have it connected by a pro at the breakout where the power delivery changes to your equipment so you can say with accuracy that the graph shows the actual power, fluctuations, distortion, sag, spikes, etc that is coming from the power co. The only evidence they will accept is from this type of recording. Then, you can find out if you are getting what they promise, or if they are out of spec.

Compare this to the spec required by your CNC. Lot of work, but you can talk to the power co until you are at screaming stage, and they won't do anything without the proof.
Edit to add: Do NOT tell the power provider you are putting in a Dranetz monitoring, they may adjust their feed if they know you are recording. I've done this successfully at several power suppliers when I could show 40-60mS spikes of 298V on a max 240V leg.

Edit again; here's a rental place for the Dranetz:
www.atecorp.com

Dranetz HDPQ Xplorer Plus Power Quality Analyzer Rentals

The Dranetz HDPQ Xplorer Plus is an exceptional power quality analyzer with superior PQ monitoring abilities in a portable and durable package. The...
www.atecorp.com www.atecorp.com
Click to expand...
Oddly the only machine giving me fits this time is the cnc plasma. The table itself is all single phase. But the plasma power source which can be a handheld unit if I had that torch is 3ph, and it gives me an input voltage error.

The power company uses RMS meters and my RMS readings are balanced and within tariff. The cheap non rms meter says there is an imbalance. This only happens when the power company is doing work somewhere upstream and has to change feeders. Its been about 9 months since this last happened.
 
Something is feeding harmonics into the network, that would be my guess. You need a THD meter to see the waveforms. Surprised the utility doesn't use one.
 
AV8R_87 said:
Something is feeding harmonics into the network, that would be my guess. You need a THD meter to see the waveforms. Surprised the utility doesn't use one.
Click to expand...
I've requested monitoring. Convincing them to use it is another story. I'm at the end of the line and nobody else nearby has any 3ph equipment to notice.

I mentioned harmonics and the EE all but ignored me. The field guy that confirmed my suspicion of being switched said that it could be harmonics but that's for an electrical engineer.
 
AV8R_87 said:
Somebody is using one phase only and trashing it.
Click to expand...
But only when my feed is switched. I got the foreman's number and was told to text him whenever I've got issues. Every time I have issues they switched my feed
 
There's really no difference in a cheap multimeter, and a true RMS meter when the waveform is a sine. The differences in the reported values are more likely a difference in the way the AC voltage is sensed and converted to a measurable unit by the analog to digital converter in the meter. An accurate Fluke voltmeter will read the same as an accurate "true RMS" voltmeter when the source of the measurement is sinusoidal. True RMS meters are used for weird waveforms to equate the power factors of the wave to something like a sine wave form.

However, the readings when taken on a 3ph circuit are important to get in sequence. And the Vll(line to line) is not as important as Vln(line to neutral). First off line to line leaves out one of the phases and may be subject to missing a power drop on the non-sensed phase. It also may mask an issue of phase shifting under load - which is highly dependent on how the phases are generated and distributed. To diagnose, it's always best to measure the Vln and make sure you are following the phases the same for each test.

Now we come to a new problem. There are power sources out there which are not true Y-phase generated but are derived phase split supplies, and they have NO neutral! Only a ground. In that case, line to line is the only measurement you can take. Find out(unlikely) if this is a "delta sourced" supply. If they come back and say it's a "Y-delta", ask if it has a neutral. The answer will usually be no.

There are six measurements to take: Vla-b, Vlb-c, Vlc-a are 120deg shifted line to line. The other six are simply Vla-n, with b and c to neutral as well. but - what I see from your measurements, I think you are not repeating the same phase to phase measurement each time.

You mention that the voltage fault is only present on one source. I would ask if one source is a delta, and if the other sources is a Y-type with a neutral. You may have an issue with a floating ground in a delta phase supply where only one 120V leg is being used for some feature of the machine. Unlike single phase 230V, it is extremely rare for a machine or some voltage requirement in a machine to use 2 of the 3 phases.

I wouldn't trust the supplier to do any metering or sensing. They have no gain to finding their voltage wrong, and the distortion outside spec. The only way I know to force the issue is quietly rent a Dranetz, have it hooked up and start recording for 48 hours. If you were in N TX I'd be glad to hook it up and configure for you. Any competent electrician can connect it, but the monitoring function setting might be a bit complex. A guide will come with the Dranezt, have it setup for voltage(of course), phase shift, and total distortion.

The other way to go is find a cheap oscilloscope on ebay or FB, and start digging in yourself. The issue here is it's usually not recording and visually you can't really see distortion. Setting up for phase shifting requires professional knowledge of triggering on the 'scope, but it can be done. An O-scope might be useful if you can reproduce the failure mode regularly while you are watching the signals in real time.
 
Last edited:
I'm not an EE or anywhere even close. But I'll bet there is something else going on, not yet discussed. I'd ask for a step back and more information.

Do you have solar?
Have you dived into the equipment manuals to determine what parameters they work within?
Are there solar generators in the area?
Are there large consumers near you?

July of one year we travelled, had big storms. Got home and HVAC systems shot. Daikin mini splits.
They were replaced under full warranty. Company suggested some people put surge protectors in line because of HVAC board sensitivities.
Within a year it happened again. They would not fix under warranty. Nobody could explain what was going on.
Troubleshooting:
Asked solar company if they had any ideas.
HVAC company and one of the top phone a friend Daikin experts had no ideas, other than to install surge suppression.
Electrician had no clue, BUT measured our line voltage as 253V or higher at times. Suggested power company was not handling things on there end, perhaps had higher voltages due to a large plant next door.
Called power company, who came out within minutes (they have to pay out when it is there fault), opened up our meter, measured each leg and it was acceptable.
Power company thought we had a grounding issue and pointed out a specific area.
Electrician said that was not the issue.
Weeks later, reading Daikin manuals, they only accept up to 240V +/- 5%, so 252 max. 253 damages them, at times we saw 256.
Talked to solar guy again, who talked to others...

The issue is that we have 3 inverters on 3 systems. Each of these reads the line voltage and adds a volt to push the extra electric back to the grid. So if line voltage in was 250, out had to be at least 253.
Solution:
We payed 17K for a new HVAC system after many days of farting around.
We installed surge protection at each outdoor unit
We installed a buck boosters transformer at each unit to drop the line voltage to something safe for the HVAC computers.

No issues since.
Nobody knew nothing!
 
Well as of right now both meters are in agreement. 246-246-248 and everything works.

I personally don't have anything solar. And don't know anyone that does. We're not in an industrial area and sort of at the end of the line. Hence why we get the high leg delta.
 
Instructions for dropping the voltage to a safe level to use a PC as an oscilloscope.

secondlifestorage.com

Turn my laptop into an oscilloscope/THD meter?

Lots of companies tell you what the total harmonic displacement of their products are, but how can we be sure they're telling the truth? I mean, companies also say their 18650 cells have 9000000mAh too. But actual testing reveals more like 500mAh. So I like to test things. Now, I've never seen...
secondlifestorage.com secondlifestorage.com
 
Racerx said:
Well as of right now both meters are in agreement. 246-246-248 and everything works.

I personally don't have anything solar. And don't know anyone that does. We're not in an industrial area and sort of at the end of the line. Hence why we get the high leg delta.
Click to expand...
Your supplier did something to regulation. Glad they sorted it out for you.
 
I'll echo what others said, best bet is Power Quality recording meter, and if at all possible record both from normal and abnormal feed. Some Utility EE's are .. . unhelpful(source: Am utility EE).
If possible who you want is the local meterman. They'd be the one to hook up and then download data from recording meter, then the power quality engineer can interpret.
Voltage is one thing, per NESC utility is allowed +/- 5% at the meter. 258 is a bit spicy, but may be that abnormal feed is higher voltage and your transformer is tapped for a typically lower voltage.
There's any number of issues that could play on a 3 phase service. If you've got 2 transformers on the pole outside it'd be an open delta service and if loaded up open delta can collapse, etc etc.
 
DreadPirateWill said:
I'll echo what others said, best bet is Power Quality recording meter, and if at all possible record both from normal and abnormal feed. Some Utility EE's are .. . unhelpful(source: Am utility EE).
If possible who you want is the local meterman. They'd be the one to hook up and then download data from recording meter, then the power quality engineer can interpret.
Voltage is one thing, per NESC utility is allowed +/- 5% at the meter. 258 is a bit spicy, but may be that abnormal feed is higher voltage and your transformer is tapped for a typically lower voltage.
There's any number of issues that could play on a 3 phase service. If you've got 2 transformers on the pole outside it'd be an open delta service and if loaded up open delta can collapse, etc etc.
Click to expand...
Appreciate that. High leg Delta with 2 transformers. Previously we did have one collapse which has been replaced. It's not necessarily that the voltage is too high. I have changed taps on my step up transy for the crane to keep that happy. That's the only 480 in the shop. My equipment has still worked when when both my meters agree with one another while voltage is high. But when they switch to an abnormal feeder, the meters no longer agree. One meter gets a line to line imbalance. Line to ground readings don't match. And my equipment takes a sick day.

I know all about what's allowed... Unfortunately. They'll look at their RMS meters and say we're within tariff gfy.
 
Happy to help. That's where the PQ is important, RMS voltage is only part of the game. Recording also very important - tough to say what station or midline equipment might be doing that could affect you.
I've seen issues with failing line caps causing THD problems, improperly grounded/bonded telecom creating interference, you name it.
Also seen customer equipment not set appropriately create issues. Had a very savvy customer give me a scope trace on when we were reading meter with a powerline carrier system. Had an uninterruptable power supply on a server buzzing when it happened. Sure enough that was it, the UPS was set more sensitively than NESC allowable so everything was ok, just wear and tear on microswitches and the annoyance of the momentary buzzing. 1:1 isolation transformer fixed it, but hummed louder than the UPS buzzing. Ended up setting the UPS to medium sensitivity and that solved it.
Also had a customer complain about intermittent RF interference(ham radio guy). Back and forth over months as we had increased voltage on a Tx line so seemed plausible we were causing it. Turned out it was happening when his neighbor fired up his Model T with a spark gap ignition - pretty good RF transmitter that. Gotta be a detective on PQ issues.
 
Back in my EE days I installed a lot of electric diffusion furnaces by Thermco for wafer fab places to provide passivation and dopants to semiconductor wafers. We had 3ph Y power needs and a lot of amps. The temp in the diffusion tubes needed to be very accurate temps. We used hi power driver boards with a basket of SCRs to trigger the power to the heating elements. We had row, after row, after row of these big furnaces and we'd put all kinds of spikes from the SCRs firing on the 3ph lines. It was so bad it would affect our controller so we had to embed a very high quality power conditioning on our own electronics. For others using the same source, woe - had a lot of problems. No way I know of to smooth the SCR spikes when they clamp down the amps and the voltage goes ballistic. Good times, at least I wasn't jumping into a hot reactor.
 
Back in the day (as it is said), before I sullied my soul and became a lawyer, I was in Tech Support for the COM Products Division of the Bell & Howell Company. We supported good stuff, big, honkin', expensive machines that took computer data from tape or IBM bus channels and printed it onto 105mm film. It was a Big Deal back then.

But clean mpower was a must, and when unexpected system halts happened, it was almost always the result of poor power, and that's when we broke out the Dranetz powerline monitor. No recording back then, but foot after foot of thermal paper printouts, from which we could tell Southern California Edison to Suck It - they had a changeover within five seconds of one AM that affected the power feed to the Security Pacific Bank operations center in Brea, and it was a doozy of a glitch. Went from "our power is pure" to "man, I never knew we were doing that" in just a few seconds. Then, they fixed it.
 
Very interesting thread. A true RMS (root mean square) volt meter works differently than a cheap meter. Cheap meters are typically average responding and RMS calibrated (.636 versus .707). A pure sine wave reads the same. If there is a difference it's usually the total harmonic distortion of the wave form i.e. it's either not a sign wave due to poor power factor or there's spikes or other crap on the line. I'd use a Dranetz or a digital recording scope or preferably both. I've seen a bunch of very expensive equipment killed in NorCal always at the same time of day - 3 in the morning when the power company switched sources.
 
OK, one more power up story. I was at UCLA, and they had a partnership with Gulf Gen Atomic looking into fusion(before I figured out it was a lost cause). Gen Atomic wanted to build a Tokamak and Gulf wanted the insurance policy for the reduction in oil/gas demand if the fusion actually worked. They built the first small one just south of Oceanside CA, and it was part of the split grid of SoCal Edison(SCE), and San Diego Gas and Elect.(SDGE). Of course, they had a tie-line and the tie line could transfer a few GW between the two in case San Onofre plant had an outage.

When we had to fire the Tokamak, we had to give both of the utils advance notice. They gave us a time frame, and it was usually a Tue from 3-430am, and we could only load the grid for 40 min. More than once, we blew up the transformers located in the valley near Oceanside. We'd brown out Leucadia, maybe Del Mar, and sometimes camp Pendelton, which got them pretty mad.

After maybe the 10th time we crashed the SoCal grid, both utilities gave us an ultimatum that we had to build a dedicated transformer plant onsite. They est the cost was going to be about $22M, and still - we could only use it at O-dark thirty. Didn't take long, the post-grad guys decided that fusion was a black hole with no chance, and they shut it all down. The Tokamak still exists, it's sitting in a glass enclosed display in Atlanta at a science museum. Sometimes I wonder how many babies were the result of our power up tests.
 
Unfortunately renting a dranetz is probably not in the cards. This issue while recurring, I have no way of knowing when. It last happened 9 months ago. Then 4 months prior to that. Usually it's only a couple hours. But it has gone on for 4 days. Luckily nothing was negatively impacted, aside from productivity. Going to try at get Comed once again to monitor. If it continues happening I'll file a utility board complaint and request it.

Since acquiring 250k worth of overhead cranes last week Im still hoping to build a new building. At which point I could ditch the high leg delta and get 480. The caterpillar dealer built next door 15 years ago had them pull 480. So that cost to pull it from Lord knows where won't be as terrible.
 
If I had that kind of huge infrastructure dependent on clean power, I'd BUY a ****ing Dranetz. monitor 7/24/365 with alarm setting and wifi or BT reporting.

Oh well, you have a solution in hand, whether you employ it or not is up to you.
 
You must log in or register to reply here.
Back
Top