SWR LA-15 Power transformer

[strebor71]

I have aquired a SWR LA-15 from a garage sale. They said it was not working but did not know any more than that. after inspection it appears that the only thing wrong is that the fused power entry is all busted up, the solder points at the base of Q2 appear to be missing some of their solder because there was a small gap between the board and the pins and for whatever reason the R33 (a 16ohm 10w resistor) was missing. I added some solder to the Q2 pins and temporarily replaced the R33 with a 100ohm cement resistor (because thats all i had here) just to see if it would get it working. Very shortly after turning the unit on the transformer becomes very very hot. After locating the schematic, I noticed that the voltage at BR1 is supposed to be 59V, the transformer in my unit is labeled as 43V. My questions are 1. is this the wrong transformer? 2. if it is the wrong transformer is that why it is getting so hot? and 3.does it have anything to do with the 100ohm resistor i used as a temp.?

I read through the other post about a la-15 tranny geting hot, but i am afraid to leave it on long enough to do any testing or bias adjustment.

here is a link to the preamp schematic
http://music-electronics-forum.com/a...a-preamp-d.pdf

[Jazz P Bass]

Do you have the schematic for the power amp section?
The preamp is most probably not the problem.
First off, the transformer 43Vac will get rectified by BR1 & bump up to 59Vdc.
If the transformer is getting hot, then either it is bad or it is being asked to supply way to much current.
You need to go a little further with the troubleshooting.
If at all possible remove the connections from the power supply board to the power amp.
(I have never worked on one of these amps)
In this state you can verify that the power supply is functioning & outputing the correct voltages.
If the supply is good, a good place to start would be the output transistors.
They need to be checked for shorts (if they are shorted they are bad)

[oc disorder]

Pwr amp schem here http://music-electronics-forum.com/t10525/

Also if you scroll down Janou put a link to some pictures of the board in a rar file. (approx 12 meg)

As Jazz P Bass suggested chek the transistors for shorts usually between collector and emitter which are (as you face the transistor) the middle and right leg.
Check both outputs on heat sink and drivers on board just in front of o/p's

At this stage R33 is possibly not important. It is a resistor in series with a capacitor that feeds the horn speaker - unless the horn is shorted which would explain the high current draw making the transformer hot.

[strebor71]

Sorry I had a little brain fart on posting that link, here is the power PCB link
http://music-electronics-forum.com/a...r-amp-rev-.pdf

I thought that might be the case with the 43VAC going to 59VDC. I have tested the transformer for continuity between the two windings and also resistance between the outputs and they all checked out ok so I powered it up not hooked to the PCB and the voltage is 43.2Vac both sides and does not get hot at all with no load. so i guess the tranny is most likely ok.
I have not yet had the Power PCB connected to the Preamp board. I wanted to see if the power PCB looked like it was working ok before i hooked it up. when I am testing voltages at points on the board, where is the best place to connect the ground lead of the multimeter? would one of the ground posts at J3 be good? what should the resistance across VR1 be, or is that something that needs to be adjusted when powered up and scoped?

Would the output transistors be Q1&3? do i need to remove them from the board to test them or can that be done still in the circuit?
oc disorder, what do you mean by "Check both outputs on heat sink and drivers on board just in front of o/p's"

I really appreciate all the help guys.

image of the board:

[olddawg]

Have you checked the power supply rectifier and the filter caps? Hook the transformer back up and float the B+/- rail voltages to the power amp. Divide and conquer. If the transformer doesn't get hot with the rails disconnected you probably have shorted output devices. BTW, don't let the transformer get too hot. It may have an inernal thermal fuse. If that pops you will need a new transformer or perform transformer surgert.

[oc disorder]

Sorry my abbreviation "both output transistors" so yes Q1 and Q3 and Q4 & Q5 being the drivers.
You can do a basic test for shorts while in circuit and better test Q3 too.
The power amp should work on its own without the speaker or pre-amp connected.
Those ground points at J3 (shown as J2 on schematic ? )would be fine.
Im sure the transformer is ok. It just doesn't like a short circuit or whatever is causing excessive current draw.
VR1 is a 1k trim pot which is probably ok just check it hasn't been damaged by a previous over enthusiastic tweaker.
I gather this is a bit touchy when the time comes to adjust the bias.

Have a look here if you want some info re testing transistors
This really applies to an out of circuit transistor (unsoldered-removed).
Testing a transistor Transistors

"Test each pair of leads both ways (six tests in total):

The base-emitter (BE) junction should behave like a diode and conduct one way only.
The base-collector (BC) junction should behave like a diode and conduct one way only.
The collector-emitter (CE) should not conduct either way.
The diagram shows how the junctions behave in an NPN transistor. The diodes are reversed in a PNP transistor but the same test procedure can be used."

Transistors Q1,2,3&4 are all NPN while Q5 is a PNP thus on that one the leads need to be reversed to test.

Check all the emitter resistors too R3 ,R4 and R9 . R 3 and 4 are very small
.1 so will probably show as a short.
While you are there lift one end of diode D1 and check that with a diode test function on meter and while its lifted up check R7 100 ohms.

There is some info on another thread regarding biasing.
http://music-electronics-forum.com/t10536/
9 times out of 10 an amp like this has been overdriven till an active component (transistor) packs in. When it does that other transistors will burn up too.
They may still look alright but with power applied further burnout happens instantaneously. Before you get to that stage I suggest you build a light bulb limiter.
Search for it on this forum. Its simply a light bulb/globe in series with the incoming mains which helps prevent further disasters !

[J M Fahey]

Fully agree with above suggestions and add:
1) pamper (meaning be careful and use a light touch) the amp parts you have, components can be bought over the counter but if the power transformer is burnt or simply the thermal fuse is triggered, it's dead for all practical purposes.
An empty board can't be bought, so unsolder carefully, do not overheat pads and tracks, do not wiggle parts when pulled, etc.
2) the light bulb limiter is essential.
3) I almost forgot, work with no speaker or other load connected until reasonably sure the amp is repaired.
4) You can start by testing the output and driver transistors for shorts , as suggested above, still unsoldered ("pamper the board").
Not as good as testing them "free" but we are looking for a gross short here, it may be from collector to emitter (center to right leg) or base to collector (left to center).
External components will *not* hide a short, even if in parallel.
If some transistor gives doubtful results in board, then remove it carefully for further testing .
This way you avoid some desoldering and pulling "just for testing"
That said, I would probably just replace Q1/3/4/5/6 because they are the most stressed parts in this situation.
Even if only 1 or 2 of them are shorted, the others are probably degraded, so it makes sense to start by replacing them all.
You will probably have other bad parts, but these are highly suspect.
5) Before you perform these tests, check the "raw" transformer (all secondary wires "in the air") as suggested, and then the unloaded power supply (transformer into bridge and main filter caps, unconnected from the rest of the amp).
Solder a 4700 ohm, 1 W (even better 2W) resistor across each big filter legs (C12/17) because when unloaded it will take forever to discharge.
Mount them neatly, I suggest you leave them there forever.
6) Consider ground all copper on PCB which joins or is in contact with C17 negative, C12 positive, J3 1/2/3 pins, J2 5/6 pins, etc.
Good luck.

[strebor71]

Thank you all for your help. I will parse through all the comments and start doing some testing. I live up in the sticks and will be making a trip near an electronics supply store this afternoon, so maybe i will just go ahead and purchase replacements for Q1/3/4/5/6 as J M Fahey suggests. And this looks like a good excuse to purchase a better iron as well. I will post my findings as soon as I finish my testing.

[Jazz P Bass]

You need to find out the fault before purchasing parts.

[strebor71]

You need to find out the fault before purchasing parts.

I had that same thought while putting together a parts list. I will wait untill i know some more. from my first quick look it appears that Q1 may be bad, it is testing differently than Q2, appears as though it doesn't matter which way i test it, continuity both directions, so that one is a cantidate for removal and more testing. I also rounded up parts i have here at the house for a bulb limiter. What wattage bulb should i use?

Hook the transformer back up and float the B+/- rail voltages to the power amp.Divide and conquer.

How do i go about doing this? just lift the legs on the two resistors R30 & 31 to isolate from the rest of the board?

I do need to get a better iron that has variable temp though. any tips on one of those that works better than the price would suggest?

[strebor71]

Update to my project (after a short delay)

OK well, I finally have an update to my project that i promised. I have received the parts I ordered for the Power board of my LA-15. I was able to find a site that had most of the original parts that i need to replace. I do have a question about The part at Q2 labeled as a T0-220 2N6474 on the schematic, however the part that is soldered on the board is a Sanken C3834. From what i can dig up it looks like it is a "Silicon NPN Triple Diffused Planar Transistor" (the C3834). Seams like it may be that someone replaced Q2 with this part maybe? are these two parts equivalent? doing some looking of my own it would seem that it should be a "NPN Bipolar Power Transistor" whats the difference between these two transistors? I have a bunch of random parts harvested from electronics over the years, is there an equivalent part i could use?

I will be replacing Q1,Q3,Q4,Q5,Q6,Q7,Q8,Q9 & R33 over the next few days, and hopefully I will know more about what to do about Q2 from all of you. Then I get to try out my new Light bulb limiter and see if all of my tinkering has paid off. Thanks again to all of you who have helped me on this.

[oc disorder]

The main difference I can see between the 2N6474 and the 2SC3834 (japanese numbering system) is the 2N6474 is rated as 4 amp collector current and the 2sc3824 is rated at 3 amps. They both seem to be general purpose replacements.
It's tricky to test the power supply by itself when its all built into the one board.
The options are to remove active components connected to the supply lines...
sometimes messy and a possibility of board/track damage or

Find clear track after the capacitors (both + and -) and using a sharp knife (think they are refered to as box cutters in US - Aust they still get called Stanley knives after the manufacturer) cut a straight line across the track and another parallel beside it about a mm away then using the hot tip of a soldering iron peel away the 1mm of track. This is severing the power from the amp and should be done in such a way that it can be easily bridged or joined after without looking too messy.
Beware of plated through holes where the track is continued the other side after passing through a plated (joined ) component hole.
Sometimes I solder limiting resistors there so the power has to pass through these to get to the amp. This enables some testing as the resistors act like fuses.
The light bulb limiter is an eaiser way to do this but at the primary end.
I mainly use 60w and 100w globes did have a big 200w one till I bumped it !
Q2 probably needs to be tested out of circuit. In the circuit there are low resistance paths around the transistor.
You probably need a solder sucker to remove the solder from the double sided board.
There is another product that here is called solderwick which is copper braid and when heat is applied to the solder via the "wick" it soaks up the solder.
Its tricky without a good iron ..I use a weller temperature controlled but I'm not sure whats available where you are.

[strebor71]

OK, so after replacing several components that i purchaced from Audio lab of Georgia and removing, testing and re-installing Q2 (2SC3834, which tested OK), I was ready to begin some testing. I plugged in my new light bulb limiter with a 25w bulb and plugged in the amp board to the limiter. After switching the power to the board on, the light bulb lit up at first and then went out, so i turned the power off fearing that the bulb had burt out. However that was not the case, the bulb was not burnt out. So, i switched the power back on and left it on. Then i started testing voltages with the negative lead connected one of the grounds on J3 to see if it was actually getting power. The transformer was not making any more noise and was not getting hot. I started with the pins on the other end of J3 marked as +-32v on the board itself. It tested at 52vDC not exactly 32 but it was getting power and the transformer wasn't melting down. I tested several points on tghe board for voltage and read several different voltages from random spots on the board. I really have no idea of what to test for besides whether or not there is power and how much. The strange thing is that after switching my multi-meter to read AC voltage I tested the same pins on J3 (the +-32). The tester read the voltage at 115vAC! So i tested several other points on the board and they also read AC voltage. How can there be both DC and AC voltage? I didn't attach the caps (4700 ohm) across the filter caps due to there price and the board continues to read nearly full power for a long time. My question is whats up with the board reading both AC and DC voltage, what would be the reason it is reading 52+ volts to pins on J3 that are supposed to be 32, or is it ok, and lastly if someone could tell me what voltages should be at a few points on the board so that i know what they should be when i test them.

the testing i have done so far was with nothing connected to the board other than the power cord.

[J M Fahey]

The strange thing is that after switching my multi-meter to read AC voltage I tested the same pins on J3 (the +-32). The tester read the voltage at 115vAC! So i tested several other points on the board and they also read AC voltage. How can there be both DC and AC voltage?

The crappy multimeter you are using "reads" DC as AC and to to make things worse, with 2 or 3 times the actual voltage. Oh well.

I didn't attach the caps (4700 ohm) across the filter caps due to there price and the board continues to read nearly full power for a long time.

You must be kidding, those 4700 ohm resistors cost 10 cents each.
EDIT:

I really have no idea of what to test for

Start by testing for shorts, using the method posted above.
Start testing in circuit, we are looking for gross shorts here.
I find it strange that your bulb limiter shows no shorts (blink and then fade off).

[strebor71]

You must be kidding, those 4700 ohm resistors cost 10 cents each.

for some reason i was thinking i needed 2watt 4700uF capacitors not resistors, and those are not cheap. I have soldered two 4799 ohm resistors to the filter caps. I am fairly certain that The circuit has no shorts and that according to the schematic, the voltages i am reading in the circuit are to be expected. The only thing left to do is hook up the pre amp board and see what happens (I will still use the bulb limiter I guess, just in case). Thanks again from those that added there comments and advise.