MEP Frequency Meter is Funny

[ETN550]

Hi everyone and good evening!

If all of the smaller MEPs use a similar method of running the frequency meter then maybe anyone can help. Please chime in. MEP 002 and 003 panels look the same as my 016B.

One of my 016B's has decided to challenge my knowledge of the frequency meter system.

I've had the gen a few months and have ran it and load tested and sorted a few other minor details. All good.

Well the other day I did a cold start at about 45 degrees and it took right off with no preheat. I usually leave the throttle in the normal running position when I hit the shutdown switch. Therefore it is already set when warmed up.

So I was surprised to see the freq meter peg at 65 with a cold engine.

I don't have a freq test meter yet but I will get one. Meanwhile I started my second set put it on 60hz and then adjusted the first set until the drone was very slowly going in and out liking synching engines on a twin aircraft.

The meter on the first set eventually krept down to about 63hz when the unit warmed up with load and after I turned the cal screw on the face of the meter which was good for about 1hz of movement.

The manual shows an adjsutment screw on the transducer but I have not had the chance to mess with it.

The meter is the old style Ideal Precision brand. Any suggestions as to why this happened? Is this a typical problem? Any fixes better than using original parts? Like a replacement meter that picks up direct off of the 120VAC outlet feed in the panel and avoids the transducer?

Please help, Thanks!

 

[Keith_J]

Digital meters can be picky, they read fine without a load but put anything other than pure resistive on them and they don't seem to deal with the back EMF.

The adjusting screw is the bias screw, it should only be moved with known frequency. It affects the entire scale. Basically, the frequency transducer generates a tiny current which is proportional to frequency. The meter's needle is connected to an electromagnet which forces against a set of tiny springs. The torque is proportional to current, same with the resultant torque of the springs. That screw adjusts the spring.

Yes, some transducers have adjustments. These should only be adjusted with an accurate meter. And some transducers can be plugged into utility power which is a good frequency reference, being between 59.9 and 60.1 Hz under normal conditions.

Finally, IF you are brave, you could use the two lightbulb method to use utility power as a reference. The bulbs flash when out of sync with mains. Being a tiny load, it is generally safe. And this is how the power unit trailer transfer switches twin lights worked.

 

[ETN550]

Digital meters can be picky, they read fine without a load but put anything other than pure resistive on them and they don't seem to deal with the back EMF.

The adjusting screw is the bias screw, it should only be moved with known frequency. It affects the entire scale. Basically, the frequency transducer generates a tiny current which is proportional to frequency. The meter's needle is connected to an electromagnet which forces against a set of tiny springs. The torque is proportional to current, same with the resultant torque of the springs. That screw adjusts the spring.

Yes, some transducers have adjustments. These should only be adjusted with an accurate meter. And some transducers can be plugged into utility power which is a good frequency reference, being between 59.9 and 60.1 Hz under normal conditions.

Finally, IF you are brave, you could use the two lightbulb method to use utility power as a reference. The bulbs flash when out of sync with mains. Being a tiny load, it is generally safe. And this is how the power unit trailer transfer switches twin lights worked.

Keith,

I was thinking of getting a freq meter anyway. Someone said the Kill a watt plug in measuring device is $25 and has freq display. Accuracy?? Multimeters with freq are around $90 and up at Home Depot.

The Freq transducer takes a two wire input off of the convenience outlet which is directly tied to the T1-T4 winding.

So what do you think about the idea of disconnecting the supply from the freq transducer and connecting household 120VAC to supply it and calibrate the meter against the measured household freq? It appears that the freq input and meter output and meter itself are totally divorced from any other generator connections, components, or wires (so no harm to anything else?) To be totally safe I could disconnect all transducer and meter wires and run new jumpers to the transducer and meter just for the household test.

Doug

 

[Isaac-1]

Get the Kill-A-Watt (make sure it is the basic model P4400, as some of the fancier ones that calculate electricity cost do not have the freq meter) they are handy to have around and do a good job on reading frequency from generators, many cheaper multimeters will give false readings on generators due to the wave form, I have one "cheap" $50 multimeter that reads about triple the actual frequency when connected to a generator, I also have a Fluke 87V that gets it right every time, but it does also MSRP for over $400 (I paid $120 or so used on ebay).

Ike

 

[BuckRyp]

Sorry wish i know more about MEPs, i have painted a few MEP 12`s just this month but i dont get to work on them. Good luck! need anyone to paint it im your man lol

 

[steelandcanvas]

I usually leave the throttle in the normal running position when I hit the shutdown switch. Therefore it is already set when warmed up...

Usually or always? I understand once you get the throttle set where it belongs, you leave it there.

 

[Keith_J]

Yes, always leave throttle (governor speed control) at 60 Hz. The voltage regulator lasts longer that way.

The basic Kill-O-Watt meters work well enough and are somewhat immune to back EMF (reactive current) which throws off meters.

Using utility current to calibrate the transducer and meter is a good idea, as long as you are certain the transducer is running off one of the windings. Since nearly all MEPs for AC are based on multiples of 120 volts, meaning each output winding is 120 volts. If using utility current, make it safe by insulating the 120 volt terminals. And never short the output, the electronics don't like that.

 

[ETN550]

Get the Kill-A-Watt (make sure it is the basic model P4400, as some of the fancier ones that calculate electricity cost do not have the freq meter) they are handy to have around and do a good job on reading frequency from generators, many cheaper multimeters will give false readings on generators due to the wave form, I have one "cheap" $50 multimeter that reads about triple the actual frequency when connected to a generator, I also have a Fluke 87V that gets it right every time, but it does also MSRP for over $400 (I paid $120 or so used on ebay).

Ike

Great info. I'm on the hunt for a p4400. I think camping world has them. I like the wattmeter too as it will be useful for both camper and generator!

 

[ETN550]

Usually or always? I understand once you get the throttle set where it belongs, you leave it there.

Yes I read about the voltage regulator not liking off scale frequencies. I totally set up my fuel governing system on both units to include lowest hand knob set point at low idle no load = 55 hz high idle no load = 65hz.

I do fine adjustment because I'm an anal nerdy engineer by degree each time I have a steady load. But yes, now, I always leave the hand knob as is when I shut down. I hope 55 to 65 is not a problem for the VR.

 

[ETN550]

Yes, always leave throttle (governor speed control) at 60 Hz. The voltage regulator lasts longer that way.

The basic Kill-O-Watt meters work well enough and are somewhat immune to back EMF (reactive current) which throws off meters.

Using utility current to calibrate the transducer and meter is a good idea, as long as you are certain the transducer is running off one of the windings. Since nearly all MEPs for AC are based on multiples of 120 volts, meaning each output winding is 120 volts. If using utility current, make it safe by insulating the 120 volt terminals. And never short the output, the electronics don't like that.

Okay, then I have my assignment! 1. Buy a Kill a Watt. 2. conduct calibration with 120VAC household. I did just read the big "Frequency Meter and Transducer" thread and gained valuable info. while you were preparing your reply! Thanks for that info too. Thanks to everyone. I'm going to reset the meter needle bias to a midpoint and hopefully just calibrate the transducer to get the meter accuracy restored. If not, then I'm back to the big thread to re-read the transducer testing methods!

 

[Keith_J]

Higher frequency isn't an issue. The problem is lower frequency since the exciter current will be low, causing the voltage regulator to peg. Running the output transistor(s) at maximum causes the issues.

The engines can take the abuse. Just don't shut down immediately after hard use, let the engine cool down. The most important is the injector system as heat makes varnish. Running at low power reduces injector temperature and the booster pump can circulate fuel though the injection pump, cooling it.

On gasoline engines, valves get hot under load. Letting them cool before shutdown prevents warping.

 

[ETN550]

I did some testing and now have some info.

I bought the Kill a watt and confirmed my house at 59.9Hz and 120VAC.

I ran the gen and the meter pegged over 65 when the set was adjusted to 60.0 on the kill a watt.

I removed the transducer and disconnected the wires from the meter.

I installed a suicide cord from the house to the transducer and jumpers from the transducer to the meter. I pulled a plastic screw plug in the transducer to reveal the calibration screw.

Well. what is happening is the system has some sort of drift issue. I was initially able to quickly and accurately move the cal screw to effect over 5 hz reduction on the meter to get it centered on 60hz.

Over the next 15 minutes or so the meter slowly kept dropping until it got to around 57Hz.

I put a hair dryer on the transducer to heat the metal housing to warm to the touch but not hot and within the next 10 minutes or so the reading fell to 56 Hz where it seemed to settle down.

Unpluging the power and re-applying the power within a minute or so and the meter resumed its pervious position of about 56.5 Hz (housing was cooling off).

I do not have a good enough hand meter to read the required 100uA at 60Hz indicater by the panel meter to verify the meter's accuracy vs transducer output.

The meter cal screw has about +/- 0.5 Hz adjustment. I returned the meter bias to the mid point to retain +/- 0.5 adjustability.

I suspect the transducer is bad and exhibiting drift which will only get worse with time. Interestingly, it came on quickly from one day just fine to the next shifted and drifted to 65Hz. Am I right to assume this drift problem is not the meter?

Is the transducer available at reasonable cost?

I do have the option of swapping the transducer on the other gen set I have to see how it might behave on the bench with household 120vac.

What do you folks think??

Doug

 

[Keith_J]

Yes, the transducer is shot. Running it open (without a resistive load) can cause this. Reasonable cost? Not likely. A 0-200 microampere 55-65 Hz transducer isn't a standard item. TransData makes a 55-65 Hz, 1 milliampere unit which you could use, employing a current dividing network to reduce the current into the meter. This is just a calibrated resistor over the terminals of the meter.

If the Kill-O-Watt works, let it be. If you must have like-stock performance, contact TransData

TransData | Power Transducers

 

[Isaac-1]

There are a number Crompton brand 60 hz 1 ma transducers on ebay at the moment with buy it now prices starting around $50, bids around $20.

 

[jbk]

delks charged me $45.00 a few mounths ago for a transducer for my 002. if i couldnt locate one i was going to use an old style reed type freq. meter. these are avaialable for $25-$30.00 check ebay. they fit the hole and require 120 volt ac.

 

[PeterD]

I did some testing and now have some info.

I bought the Kill a watt and confirmed my house at 59.9Hz and 120VAC.

I ran the gen and the meter pegged over 65 when the set was adjusted to 60.0 on the kill a watt.

I removed the transducer and disconnected the wires from the meter.

I installed a suicide cord from the house to the transducer and jumpers from the transducer to the meter. I pulled a plastic screw plug in the transducer to reveal the calibration screw.

Well. what is happening is the system has some sort of drift issue. I was initially able to quickly and accurately move the cal screw to effect over 5 hz reduction on the meter to get it centered on 60hz.

Over the next 15 minutes or so the meter slowly kept dropping until it got to around 57Hz.

I put a hair dryer on the transducer to heat the metal housing to warm to the touch but not hot and within the next 10 minutes or so the reading fell to 56 Hz where it seemed to settle down.

Unpluging the power and re-applying the power within a minute or so and the meter resumed its pervious position of about 56.5 Hz (housing was cooling off).

I do not have a good enough hand meter to read the required 100uA at 60Hz indicater by the panel meter to verify the meter's accuracy vs transducer output.

The meter cal screw has about +/- 0.5 Hz adjustment. I returned the meter bias to the mid point to retain +/- 0.5 adjustability.

I suspect the transducer is bad and exhibiting drift which will only get worse with time. Interestingly, it came on quickly from one day just fine to the next shifted and drifted to 65Hz. Am I right to assume this drift problem is not the meter?

Is the transducer available at reasonable cost?

I do have the option of swapping the transducer on the other gen set I have to see how it might behave on the bench with household 120vac.

What do you folks think??

Doug

First, your home's line power from the utility is 60.0 Hz. Absolutely, exactly 60.0 Hz. (there are technical reasons why, not important here...) What you see when you read 59.9 is the old +/- one count error that many digital devices give. It is not a problem, just do not think that a consumer power analysis device is more accurate than the line frequency!

Second is that though we'd love to have exactly 60 Hz from our generators, the bottom line is that except for clocks (such as the VCR, the microwave, etc.) most all home devices do not care if the frequency is off by a bit. In fact, 10% is generally acceptable, and many non-US origin devices (imported stuff) is designed to work on either 50 or 60 Hz. Of course if that device has a clock in it, there may be an internal jumper to allow the clock to adjust to the different frequencies.

Now, to the problem! (ah, ha, this moron has a suggestion!) Does your transducer have the two adjustments on the bottom? If yes, then mark the current position of each, and then turn them about ten times from minimum to maximum, and finally stop at the marks. I found at least one transducer that these adjustments were causing accuracy problems. Once you do this, then see if you can get (when connected to the house's 120 power) 60.0 Hz (or as close as possible). If you can, re-run your tests over time and temp. If not, gently tweak the adjustments on the transducer until you are at 60 Hz. If you can get the 60 Hz, again, re-test everything.

Of course you are a bit fubar'ed if the transducer doesn't have the adjustments. Personally, I'd look for a digital replacement, which may be much more reliable.

 

[ETN550]

First, your home's line power from the utility is 60.0 Hz. Absolutely, exactly 60.0 Hz. (there are technical reasons why, not important here...) What you see when you read 59.9 is the old +/- one count error that many digital devices give. It is not a problem, just do not think that a consumer power analysis device is more accurate than the line frequency!

Second is that though we'd love to have exactly 60 Hz from our generators, the bottom line is that except for clocks (such as the VCR, the microwave, etc.) most all home devices do not care if the frequency is off by a bit. In fact, 10% is generally acceptable, and many non-US origin devices (imported stuff) is designed to work on either 50 or 60 Hz. Of course if that device has a clock in it, there may be an internal jumper to allow the clock to adjust to the different frequencies.

Now, to the problem! (ah, ha, this moron has a suggestion!) Does your transducer have the two adjustments on the bottom? If yes, then mark the current position of each, and then turn them about ten times from minimum to maximum, and finally stop at the marks. I found at least one transducer that these adjustments were causing accuracy problems. Once you do this, then see if you can get (when connected to the house's 120 power) 60.0 Hz (or as close as possible). If you can, re-run your tests over time and temp. If not, gently tweak the adjustments on the transducer until you are at 60 Hz. If you can get the 60 Hz, again, re-test everything.

Of course you are a bit fubar'ed if the transducer doesn't have the adjustments. Personally, I'd look for a digital replacement, which may be much more reliable.

The transducer has one cal screw accessable through a plastic screw plug in the housing. It appears to have +/- 5Hz adjustability. In other words when the transducer settles down to 60 Hz and quits creeping I can move the needle full scale.

The drift over 20 - 30 min is the problem. It eventually settles down. The transducer "box" is two pieces of sheet metal riveted together. Is there anything inside that can be cut out and replaced?

 

[ETN550]

delks charged me $45.00 a few mounths ago for a transducer for my 002. if i couldnt locate one i was going to use an old style reed type freq. meter. these are avaialable for $25-$30.00 check ebay. they fit the hole and require 120 volt ac.

I'll check Delks. Thanks for the lead.

 

[ETN550]

Yes, the transducer is shot. Running it open (without a resistive load) can cause this. Reasonable cost? Not likely. A 0-200 microampere 55-65 Hz transducer isn't a standard item. TransData makes a 55-65 Hz, 1 milliampere unit which you could use, employing a current dividing network to reduce the current into the meter. This is just a calibrated resistor over the terminals of the meter.

If the Kill-O-Watt works, let it be. If you must have like-stock performance, contact TransData

TransData | Power Transducers

Yeah, I kind of figured you would say it is shot. It was never run open. It is strictly a drift problem. I did see some 55 - 65 1MA transducers but they were too large to fit in the box and very expensive. A resister across the meter is not a problem if I did find a 1 MA provided temperature swings would not mess it up.

I know most electrical end devices don't care so much about the Hz but I do want the engine to be running at speed. Any reduction in speed is a proportional reduction in available power at the lower speed. Not exactly proportional but pretty close for small speed variations.

Finding an old style reed meter is also a good option, especially if it has the bottom of the meter glass is blocked off like the existing Ideal Precision gages on the unit. I would like to keep it stock appearing, if practical. The meter is handy for checking governor settings, min speed, max speed, and droop so I do not want to loose the meter altogether. Do the modern small square meters on e-bay require a transducer or are they just connected to 120vac?

 

[jbk]

i use the killawatt and hardy digital meters for testing and setting up. i dont rely to much on the panel meters, but do like them to work. most of the analog meters are just not going to be very accurate.