Strange readings............

[George A]

I have a military 1800 rpm, 2500 watt, 4 pole, 117 volt AC generator being driven by my Vidhata 6/1. I bolted a steel box to the generator top and equipped it with an analog 0-150 volt voltmeter, an analog 0-30 amp ammeter, a reed type frequency meter, circuit breaker and two outlets. I went entirely through the generator installing new bearings, cleaned up the armature and brush rigging, then assembled everything exactly as it was when I pulled it apart. It has a two brush exciter and four brushes for output....a pretty straightforward and heavy duty design.

When running, I get a fairly stable voltage (a little fluctuation from the power impulses) and can draw up to 22 amps....the full output according to the tag on the generator (there are still some governor issues but I'll sort those out in the near future). The reed type meter is pretty much useless though, as the vibration causes all the reeds to vibrate at once.....so I bought a Kill a Watt and plugged that in. That was an eye opener! I finally got some firm information as follows:

The voltage is steadier on the Kill a Watt.....however that may be due to some sort of circuitry inside that averages out the reading.

The frequency, which is really what I wanted to know, is strange. I had the engine speed set so that the panel meter and Kill a Watt read 125 volts no load, but at that voltage my frequency is around 54! If I adjust the governor so that it generates 135 volts, I can finally get the frequency up to 59!

I used two electric heaters for a load and the voltage drops from 135 down to 117 with one heater, then down to 108-110 with both heaters at once. Pretty wide spread, but that's the governor issue. I didn't think to note the cycles per second when loaded but I bet it's low.

I haven't actually checked the generator shaft speed, but I'm reasonably sure it's pretty close to 1800 rpm (higher at no load) as my pulley ratios are correct.

I'm a little puzzled, as I always assumed that the cycles per second were dictated by the rpm, as long as the voltage was correct. Why the discrepancy? Or would that be normal for a military field unit?

[dieselgman]

George,

You are right about several aspects of your setup... Firstly: Frequency and rotational speed will always be in lock-step. You want to start any adjustments by bringing the set up to 60Hz (or 50Hz outside of North America). Secondly, you want to mount your old reed meter on rubber, or better yet, completely isolated from generator vibrations. As you have noted, the meter responds to and attempts to synchronize with the AC cycles per second but cannot settle when subjected to excessive vibration. Your Kill-A-Watt does indeed average readings and is not good at picking up subtle or instantaneous changes - however, it is fine for getting your set adjusted and a nice tool for instrumenting and monitoring your output.

Voltage output... another matter altogether - not linked to frequency (although certainly affected by rotational speed). There are many various and asundry voltage regulation and adjustment schemes in use on different equipment. Modern equipment uses electronic AVR controls (in simple terms) to vary the DC excitation voltage based on output voltage and thereby stabilize and control output voltage. Your unit may use capacitor/s, transformer/s, inherent winding values, or other variations to set up the output voltage. We would need further details on your unit to help with the adjustment and interpretation of what yours is doing right now.

Your voltage output range sounds excessive... most modern stuff is within quite close tolerances of less than 5% variance. It is also worth noting here that the frequency issue is not critical for all applications... most lighting and heating loads do not care about frequency at all. Running your inverter and battery charger, electronics and such, electric motors can be very sensitive to correctly controlled input frequency.

dieselgman

[George A]

Thanks for your input dieselgman!

I completely forgot to mention that there are a couple of .1 ohm capacitors tied from the output to the generator frame inside. The originals were in pretty sorry shape, so I replaced both with the same value but higher voltage. The originals were so badly mangled that I had to use a magnifying glass to get the value of off one of them and as far as I'm aware they're correct. If I'm wrong though, that very well may be the problem. Any advice out there?

So far, I've only run a couple of heaters, and electric drill and a halogen worklight off this unit and they were fine. With such goofy Hz though, I'd be a little afraid to plug my new refrigerator in as it probably has some sort of mini computer inside.

Would some sort of an "afterthought" AVR be available for a generator as old as this one? Or even useful? Failing any fix, I might just be looking at an ST unit replacement.......

[dieselgman]

George... we would need to know a lot more about identification of your particular generator head to be of much help. Post pictures, list out the data plate information... something along those lines.

Capacitors are never rated in ohms (but rather microfarads uf), and indeed sized for specific voltage ranges... yours may be related to voltage regulation but also may have something to do with rf noise suppression on your outputs. Ohms are the ratings given to resistors... we must clarify which actual components you are handling and replacing.

I agree, that I would not try to run any modern or sophisticated/expensive equipment on the set until the frequency issue is fully resolved and under control (stable).

dieselgman

[Quinnf]

I set mine up at 62-63 rpm at no load, and it drops to 59-60 at full load, so I think your governor linkage might be sticking somewhere.  Mount the frequency meter on the wall and it will work fine.  Better yet, eBay has some very nice analog frequency meters for very little money and they're far easier to read than the reed-type, but nowhere near as "cool."  http://www.ebay.com/itm/45-65-Hz-ANALOG-Panel-FREQUENCY-Meter-HERTZ-Indicator-/271234215492

[George A]

Ohms??? Why did I type ohms?? I meant microfarads and I'll pull the inspection cover to verify their connections. I think they are for noise suppression though.

Governor linkage has been bushed, adjusted and shows no evidence of binding. Very smooth, positive action throughout its range. I'm pretty sure it's the old spring tension issue mentioned in other posts.

I'll get all the info off the generator tag and post it tomorrow. Thanks guys.

[George A]

UPDATE:

Information on the generator tag is:

Kurz & Root
volts 120
amps 22
ph 1
type A
FR 608
KVA 2.5
temp rise 60*
time cont.
serial number 95266 (this one is a bit hard to read)

Further update: I took the "too soft" spring down to our local Ace Hardware and rummaged through their selection (not much). I compared all the like diameter/length springs by pulling them gently and checking it against the feel of the "too soft" spring. Not very scientific, but I finally found one that was a little stiffer. Came back home and installed it on the engine then fired it up and started plugging in heaters. The response was better, but not much. It then occurred to me that I had assumed the heaters were 1500 and 1000 watts. One tag did say 1500 and I found the other on the BOTTOM of the heater..........also 1500 watts! I had forgotten to verify the load before any testing and was demanding 500 watts more than the generator could produce.


Engine is running at 610 rpm (7 3/4" generator pulley) and I'm at approximately 4000 ft. altitude, so I suspect it's putting out approximately 5 h.p.

I backed off the governor, started the engine and then reset it....

Set the Hz at 61.5 and get 137 volts at that setting. No load, engine running comfortably, not hunting.

Plugged in the 1500 watt heater. Got 59.0 Hz and 125.4 volts (this is over half the generator's rated capacity). Engine working a bit.

Plugged in the second heater set @ 1000 watts. Got 56.5 and 115.5 volts (full rated capacity of generator, 22 amps on meter). Engine working harder, but not much.

So, my total voltage drop is 21.5 and the Hz drops 5 cycles per second. I did have to "average" the readings a bit as they fluctuated on the Kill a Watt...but I didn't cheat! However, you'll notice that with a load the voltage is between 115 and 125, while the Hz is between 56.5 and 59.0.

May not be close to ideal, but it's a lot better than yesterday! Moral to the story: Make a checklist and follow it nextime........