Exciter voltage and adjustment on AEAD-200LE serial kb048669

[Aeronca41]

Outstanding! Glad to hear of the success.

[[email protected]]

After a bit more tinkering it developed the idle mix was misadjusted. To make matters worse, there's
no air seal on the idle mix screw, adding indeterminate amounts of air to the idle circuit. It wasn't
hard to set up a crude seal with a greased o-ring and flat washer under the lock spring. It's not perfect
but can't be any worse than no seal at all.

With a better mix setting the machine no longer hunts at no load and sags about 3 Hz from no
load to full load.

Thanks for all your help, I'm going to declare victory and retreat!

bob prohaska

[duaneb55]

I'm not surprised that you're only seeing about 1/3 total travel of the throttle with a 5K auxiliary load given that a 5K output only requires between 7-9 input HP.

I say 7-9HP as it depends on what efficiency factor one uses to calculate the power requirement.

[[email protected]]

It's tempting to think of the problem as overspeed at no load, rather than
underspeed at full load, given that the throttle is only moving about 1/3rd
of full travel at nearly 5kW (not 4 as I wrote earlier).

There's no hint the machine ever had a spring such as yours and I can't
find any documents to suggest it. Unfortunately the ID sticker on the engine
is illegible, the only number I can find is 3773804 stamped into the block
where the points box used to be.

That the spring I tried didn't help isn't too surprising; the ID was rather close
to the OD of the link and it was practically impossible to keep it from rubbing on
the rod given the offset in the anchor hole on the throttle crank.

Radial play in the throttle shaft bearing is about .010 in the vertical direction,
probably similar or a little less horizontally. I know that sloppy throttle bearings
cause trouble, especially at low speeds/loads, in cars. When the engine needs
to run within a hundred RPM of setpoint maybe .010" is too much. New carbs are
pricey, but available. Maybe they can be re-bushed.

Now that the voltage problem is fixed the machine looks usable for keeping
a fridge running, which is probably the most critical job it'll ever do and hopefully
not often nor long!

bob prohasa

[duaneb55]

Your additional speed drop is puzzling.

Note that the buffer spring goes over the link and connects to the throttle and governor levers such that it is in essence one in the same with the link and moves with it rather than create any friction or resistance against it.

[[email protected]]

The governor arm link on mine lacks the reverse bend to provide clearance near the spring
anchor hole in the throttle crank. I tried using a very soft B&S governor spring in a fashion
similar to your setup, but the added friction (I think) caused more trouble than it solved. The
speed drop went from 5 Hz to almost 10 Hz. The spring was scrubbing the rod and it was very
difficult to get it bent so it didn't foul on either the rod or the crank.

In the meantime I looked at the plugs (not new, but not awful), the compression (130/137 psi,
too good to be true) and the throttle shaft play (detectable but not outrageous).

It might be possible to use a pair of small Heim joints and some allthread to make a zero-play
link. Whether the Heim joints will have low enough friction is far from certain. The carb could be
cleaned, but the engine gives no hint of carb trouble.

One thing I haven't looked at is the valve clearance. Given the compression it seems unlikely
to be far off, but how common is valve trouble on these machines?

Thank you!

bob prohaska

[duaneb55]

There's perceptible play in the linkage between governor lever and throttle shaft. Is that worth
removing? Some lawnmower engines have "anti rattle springs" in the governor linkage for that
purpose, it might not be hard to do something similar on the Legend.

Actually there should be one there. This is on my LP/NG fueled Bobcat but the same linkage arrangement as the gasoline carburetor version.

[[email protected]]

The excessive drop in frequency (RPM) can be the result of a few things I can think of off the top of my bald head.
1) Weak (reduced spring rate) primary governor spring.
2) Governor sensitivity (hole position in governor lever) setting. More sensitive closer to governor shaft.
3) Poor engine condition.

It looks as if things behave much better with both springs in their 3rd hole from the governor shaft.
With 65 Hz no-load the machine holds 60 Hz and 120 V at roughly 4 KW. I'll test more thoroughly
to make sure, but the worst problems are fixed.

There's perceptible play in the linkage between governor lever and throttle shaft. Is that worth
removing? Some lawnmower engines have "anti rattle springs" in the governor linkage for that
purpose, it might not be hard to do something similar on the Legend.

With my thanks,

bob prohaska

[duaneb55]

The excessive drop in frequency (RPM) can be the result of a few things I can think of off the top of my bald head.
1) Weak (reduced spring rate) primary governor spring.
2) Governor sensitivity (hole position in governor lever) setting. More sensitive closer to governor shaft.
3) Poor engine condition.

[[email protected]]

Replacing R70 and the zener diode seems to have brought the 15 volt supply
back. The duplex outlets are about 120 V at 62-63 Hz and sag to 116 under a
1 KW load. The frequency drops to 55-56 Hz, which seems rather a large change.
The throttle is just barely off the stop screw, so there seems to be a lot of headroom.

It seems as if the regulator is now resisting load-related voltage changes but it
has seemingly minimal effect on speed-induced voltage changes. Pushing the
engine speed up or down with a finger on the governor produced roughly a
1V/Hz voltage change.
.
Does this suggest some subtle malfunction remaining in the regulator?

Resistors R2 and R4 are still shorted out of the exciter circuit entirely. Is there a
serious risk of burning something expensive (like the stator or rotor) up if the
exciter goes back to being "its old self"? At this point I don't think that's a physical
possibility, but I've been wrong about this machine enough times before........

Many thanks for all your help, and any further counsel.

bob prohaska

[duaneb55]

Any suggestions on how to remove the conformal coating without doing damage?

I use Brake-Kleen, a fine SS wire brush and blow it off with compressed air. Acetone will likely work as well as it is one of the main ingredients of Brake-Kleen.

Be advised that the Brake-Kleen/Acetone may compromise certain component coatings/coverings if left in contact with them too long.

[[email protected]]

Check the diodes and caps.

Excellent call. It looks like D51 (27 volt zener) is shorted. How that happened is
the next mystery; DC input should only be about 22 V at generate speed, zero
at weld speed. The caps aren't leaking or bulged.

Any suggestions on how to remove the conformal coating without doing damage?

Thank you!

bob prohaska

[Aeronca41]

Since the machine is so old, I wonder if Miller would give you a tech manual with troubleshooting info? I think I'd try asking them. tel:920-735-4356


Actually, with the board diagrams already in the manual, there may not be much additional info in the Tech Manual, but I'd ask anyway.

Agree that a toasted R70 is most likely a symptom. I would check the input diodes-if one shorted, it would have allowed AC onto those old electrolytic caps, which would cook R70-and perhaps then opened a diode as the failure unfolded. Or it could be that the old caps just started to leak a lot and cooked it. Check the diodes and caps.

[[email protected]]

It looks as if the generator works but the regulator is insensitive. The regulator is alive, the voltage across wires
26 and 25 increases from around 5 volts to around 17 as the 800 watt load is applied. It's just not enough.

It looks as if this is badly wrong. On close inspection the voltage regulator board seems to have lost its
15 volt power supply. The apparent cause is the overheated R70, which lowers the input voltage to
the regulator to 10 volts. Obviously, a 7815 can't make 15 volts out of 10. No other components
looks like they got hot and the soldering looks good, with no obvious cracking or dewets.

I'm very tempted to just replace R70 and see what happens. I admit that this is treating the
symptom, rather than the disease. Without a better understanding of the SCR side of the
circuit the disease is apt to evade me.

It seems fairly clear that the two op amps deliver an error signal to R52, but how that interacts
the the SCR section isn't at all clear to me. Can anybody point me to a description?

Thanks for reading and any comments,

bob prohaska

[[email protected]]

With R4 and R2 bypassed completely the
AC weld voltage is 79 VAC
100 Hz outlet is 131 VAC no load at 99.5 Hz
Exciter voltage is about 140 VAC

At 60 Hz the duplex outlets are at 120 VAC but drop to about 99-101 VAC with an 800 watt load at around 59 Hz.
The exciter is at around 60-65 VAC and seems to wander a bit.

It looks as if the generator works but the regulator is insensitive. The regulator is alive, the voltage across wires
26 and 25 increases from around 5 volts to around 17 as the 800 watt load is applied. It's just not enough.

Removing R2 and R4 seems like a rather drastic adjustment. Both were close to midrange and don't look like
they've been disturbed from factory settings.

Thanks for reading, any counsel appreciated

bob prohaska