Lambda LPD-422A-FM Runaway Voltage

[USMC_Spike]

It makes no sense to start another thread about this, so I hope
no one minds if I continue this post.

I guess this is the time where I say the ME TOO!

I had a working unit like this for awhile now.
Until today.  I've got a scooter with two 12V batteries wired in series.
for 24 volts DC.  It just so happens the connector for the batteries
is for the end points.  And it has terminals for the + - series connection.

As my batter charger for the scooter fried, I was going to
try and charge these batteries from my Lambda LPD-422a-FM.
I planned to use a higher voltage and current limit the batteries.
The batteries are 10Ahr.

You can figure out the rest.   I put the power leads from the
Lambda into the opposite polarity of the batteries.
When I did this I hear the tiniest of pop noises from the power
supply.

Then the faint smell of the sacred smoke...but no smoke seen.

I looked at the connections and and made them correct.
then I changed the leads to the other channel for the lambda.

Switched on the power and saw that the smoked channel was
running full out, both the Voltage and the Ammeter were
far off to the right...in Excess of full operating voltage.

I checked the pots but they had no affect either full CCW or CW,
the meters were past the highest of each meter.  So I powered
off.  I found this thread and got the manual found the schematic
and the trouble shooting pages.

R30a, and R30b measure just fine, on both A and B sides.

So the next obvious choice might be The diode CR23 or CR24.  Note 11 doesn't help much.

Checked that diode, it conducts in both directions...in circuit.
Then again so do the other two of the same type.  But in different values.
Well I guess it's time to start desoldering and testing.  So much for an easy
and quick fix.

[macboy]

Any good power supply has a diode across the output terminals, reverse biased. This does two things: it helps protect the circuit when someone connects a battery backwards (sound familiar?), and it may also conduct in some conditions when the power supply is connected in series with another power supply.

When you connected the battery backwards, this diode conducted a large amount of current. A 10 Ah battery is quite large. A smaller battery might have only caused the diode to fail short, but this one might have caused it to fail open. Then the voltage of the battery would be present, in reverse, across the output terminals.

At the very least, find this diode and replace it (note polarity, the cathode band will be at the + terminal). Other devices may have been damaged. The output transistor may be damaged if the voltage was relatively high (causing V_CEMAX or V_BEMAX to be exceeded). If that is the case, check the components driving that transistor (maybe another transistor or the output of an opamp). The output will likely have an electrolytic capacitor across it, and since these are polarized it may be damaged by the reverse voltage, but not likely.

[romany]

Not sure of the value of starting another thread just to comment.   This power supply does indeed have a diode across the 2 output terminals (actually 2 - one for the A side and one for the B side).   I managed to connect a 40V lithium battery pack backwards - on both sides - and the surge of current roasted  the lan leading to the protection diode - while leaving the diode to fight again another day.   Bypassing the burt lan fixed the problem.   To hopefully prevent this from happening again I'm installing a low amp fuse on both the A and B sides....

[coppercone2]

I broke a lambda with a battery too, they don't like batteries.

It shorted the regulation transistor and put out maximum voltage across the output. This happens because of something during turn off I think. So you need to disconnect the battery before you turn off the unit.