Trace Inverter - Low output voltage

[Timkar]

Have checked the archives but can't find anything related to my problem. I have a U2512SB Trace Inverter which was working fine. Turned it on today and output voltage is between 78 and 80 Volts AC. Batteries are fully charged and are at 13.2 VDC.
I removed unit from coach to check on bench and hook up another battery and I get same results.  I have checked web and cannot find
any further troubleshooting ideas . Anyone had this problem before or have any idea what problem may be?
TIA....Tim

[junkman42]

This will sound stupid but here goes anyways!  Is the voltmeter one that You have used to look at the inverter output before?  Some meters do not like modified sinewave!  If the unit You have is pure sinewave ignore Me I am old and feeble!  John L

[Len Silva]

I think John has the answer.  Have you checked it with that same voltmeter in the past with different results. If you don't have a true RMS meter, you will not get accurate readings.

Connect a lamp to it and compare the brightness to a similar lamp on household power.  You probably don't have any problem.

[Timkar]

Meter is a fluke 77 which I have used before. While the unit was installed I had none of the house systems operating in the bus until I plugged in shore power or ran gen set. 
You guys were right...Fluke 77 is not true RMS. Wired and plugged in a small compressor and it started up although ran slowly. Voltage went up to 114 while running unit. Problem must be further downstream. Saved me some money which I can't afford to spend right now.... 
Thanks...Tim

[Lin]

I have the same inverter.  It does not do well on things like compressors.  There is also a setting on the inverter for the threshold demand it needs to sense in order to kick on.  I leave mine set at the minimum so that even a light will turn it on.  If there is no load to turn the inverter on, you can not test the voltage.

[bevans6]

Speaking purely theoretically and because this is an interesting problem, if you have an AC meter it may read RMS AC for a true sine wave, but reading RMS value (the "equivalent power to DC" value) of a non-sinusoidal waveform is more complex.  Back in the day we used a very expensive meter that basically had a heating element in it,  and the meter read the heat produced and compared it to the heat that a pure DC waveform would produce - no doubt there are other ways to skin that cat now.  A normal multi-meter will just do a kind of averaging conversion of greater or lesser sophistication.  For a pure sine wave the peak voltage times .707 is the RMS DC equivalent voltage.  The average voltage of a sinewave times 1.11 is the DC equivalent or RMS voltage.  (Man, electronics school was 35 years ago and these conversion factors stick in there for some reason).

Now the interesting part.  An AC motor is both a user of power and a generator of power.  While running it produces a counter electromotive force (CEMF) that is in exact opposition to the supply voltage.  That means that while it is running on the non-sinusoidal waveform produced by the inverter, it is producing a pure sinusoidal CEMF in opposition to the supply.  This will have the effect of radically smoothing out the waveform, and the effect should be that your meter will now read much closer to the correct voltage equivalent of what the inverter is putting out.

I think, anyway...     

Brian

[junkman42]

Brian, think 35 years is bad try 1959!  I worked in air force PMEL before retiring in 1980.  The meters with thermal converters are still used today funny enough.  The numbers that were in the post is what made Me think about the meter thing!  I think that number 1.11 is incorrect but I doubt it really makes a rats a""""s>  LOL  take care, John L.

[Gerry H]

It's 1.4141 not that it matters, I can't prove it 

[bevans6]

The average voltage of a pure sine wave signal is 0.636 of peak voltage.  The "DC equivalent" voltage, or the  RMS voltage, is .707 of the peak voltage.  That means that to convert average voltage to RMS voltage, you multiply the average by 1.11.  1.414 is the inverse of 0.707, so to convert RMS voltage to peak AC voltage, you multiply the RMS figure by 1.414.

I've gotten old, fat, bald and slow, but i ain't gotten stupid.  I also appear to not have forgotten electric circuits 101, for some daft reason.  Plus I double checked on Wiki...

brian

[Gerry H]

I apologize for the inaccurate number. The last thing this forum needs is inaccurate information. I use that number (1.414) quite often working on audio power amps with a scope and I posted it without thinking through the original meter problem. So just put me in the old, bald, fat, slow and stupid column, but I'm still Minnesota nice.

[bevans6]

Gerry, I am baffled as to why I remember those conversion factors, since I haven't used them at all for many many  years.  They just stick in there for some reason, taking up space I could use to remember things like, oh, my wife's birthday...   I also remember the NATO phonetic alphabet and I can spell words as "whiskey oscar romeo delta sierra" as fast as I can say the words, and I haven't used that since I was 20 years old and in the Artillery.

Brian

[artvonne]

Gerry, I am baffled as to why I remember those conversion factors, since I haven't used them at all for many many  years. 
Brian

  I was at the Bobcat dealer yesterday discussing some issues with a machine I own, and remembered a part I needed. I had called about it last week, and had written down the part number, but didnt have it with. Surprisingly I remebered the part number it and rattled it off. I even knew the price.

  Our minds are incredible at remebering things that are virtually meaningless, and astounding at thier ability of forgetting the things that should be much more important.

  This discussion is really interesting, I never knew you needed some complex meter to read AC power like it being discussed. And I own a fluke 77 as well. I assume this meter reads residential line power, as well as generator power, correctly, and that it is just inverter power where we would run into this problem???

[bevans6]

What it is about meters is that they read AC voltage and report it in one of three different ways.  That could  be peak voltage (kind of not used as a rule), inferred DC equivalent voltage, also known as RMS voltage, or measured RMS voltage.

Up to the  advent of computer based meters the technology to actually measure RMS voltage was kind of clunky.  I referred to a lab meter that we had when I was in school that used a thermocouple to measure heat and calculate the DC equivalent voltage.  There are new meters like a Fluke 289 that sample the waveform and calculate the equivalent DC voltage, but as far as I know the Fluke 77 does not.  The sampling meters have a maximum frequency that they can measure, but household electricity is no problem.

An analog meter just measures the average voltage after rectifying and has an appropriate scale to read RMS voltage.  It can only be truly accurate if the signal is an accurate sine wave, or any other waveform that the meter has a scale to read.  You could have a scale for a square wave, or for a sawtooth, or whatever - the meter reads the average, and the needle goes to a certain place, and whatever number is on the scale is your reading.

A typical DVM that is not  true RMS works exactly the same way.  It reads the average rectified value and assigns a number to that, typically based on a pure sine wave, and reports that.  Since it's reading average voltage, and assuming that it is measuring a sine wave, any other waveform will get some random number.

I would personally like someone to comment or confirm my idea that running an induction motor from a modified sinewave inverter will create a CEMF that will force a sine waveform over top of the modified sine wave.  It would make measuring the voltage of  a modified sine waveform somewhat interesting, be fun to look at with a 'scope.

Brian

[Len Silva]

Some of you old farts have better memories than I, although the .707 was still in there somewhere

At any rate, a reading of 70-90 volts, depending on the quality of the stepped wave is about what I would expect from the typical inverter.

Brian, the induction motor experiment makes sense, I'd like to see the results myself.

What would be the effect of using a full wave rectifier and measuring the resultant DC,  I can't wrap my poor old brain around that either.

Golden age my @$# .

[bevans6]

If you use a full wave rectifier on a typical AC sinewave, you get .707 of the peak voltage less losses in the diodes, which I forget what they are and it depends on what kind of full wave rectifier you use.

I think.    Pride prevents me from checking until after I hit Post!

Edit:  I was, surprising the heck out of me to be honest, right!  If I was this good at electronics 35 years ago maybe I would have had an honest job instead of being in sales all my life!   

Brian