Newbie needing help with amplifier problem

[Keyhall]

Hi all, I'm new to the site and and this is my first post/problem and am trying to teach myself electronics in order to revive faulty home audio equipment as a hobby. I would say I have good basic knowledge of electronics and have already used various resources to self teach myself but have come a bit stuck with my current project.
Previously I have repaired units by visually checking components and removing them from the pcb in order to check they are functioning correctly but my current project has no obvious signs of faulty components and so am going down a new route which is to track the voltages accross the board to find the problem. So here goes...

The amp is a Linn Majik-I and it works fine for about 5 minutes when it cuts the sound off. If I leave the amp turned on it never resumes the sound but if I turn it off for a minute or two and then back on, the sound comes back on again for another minute or two.
To me this points to an overheating component issue. but what do I know

The power and display stays on so don't think it's a power supply issue
The amp does have both over current protection and over temp protection but if the system over temps it is supposed to adjust the volume by - 1db at at time and keep attempting to resume the sound, it does not do this and I have checked the temp of the thermistor and tested this and all looks ok. (it is a 33k thermistor which reads 33kohms at room temp).
I have checked the transformer outputs both before and after cutout and the voltages stay the same, as do the main Opamp chips Q301 & Q302.

Today I have started to look at the transistors on the board and have found a possible clue as to the problem but dont know why this is happening.
testing U301 U302 & U303 I get the following approx readings both before and after the sound cutoff:

U301
in 39v
out 25v
adj 24v

U302
in -39v
out -24v
adj -24v

U303
in 25v
out 14v
adj 14v

but U304 has different reading before and after the sound cuts off

before after
in -25v -25v
out -8v -3v
adj -7v -3v

is this a sign of a faulty transistor or is something before this stage having trouble?

Ill upload the schematics in a mo when I work out how to

Kind regards

Key

 

[Krapoman]

Hi,
I m a newby as well.
And a beginner in electronic.
The schematic tells Vout of the regulator U304 should be -15v.
So Vadj should be around -14v.
But Vadj seems to be controlled by Q307.
If i'm right, you should read 1,4v at the base of Q307, not less..

 

[Keyhall]

Hi Krapoman, thanks for getting back, ill take a look when I get back from work later today, the Q307 is a small PNP Transistor which I haven't checked yet.
Cheers
Key

 

[73's de Edd]

Sires . . . . . . .

MUSINGS:

1 . . .but U304 has different reading before and after the sound cuts off

2 . . .before after
in -25v -25v
out -8v -3v
adj -7v -3v

3 . . .But Vadj seems to be controlled by Q307.
If i'm right, you should read 1,4v at the base of Q307, not less..

Would you now like those technological rationales explained. . . . .?

But first, would you interchange the U302 and U304 parts positions, as it is, the
U302 is having the least power demanding circuit application, of those two..

We want to initially confirm that you don't have a component related problem,
which I don't suspect, but you have an easy answering of that , with a swap.

Then see if that NEGATIVE 15V supply still has that halving of voltage, with a bit
of run time .


73's de Edd

 

[Keyhall]

Hi, thanks again for the replies,

Krapoman
If i'm right, you should read 1,4v at the base of Q307, not less..

Q307 before/after sound cutt off (taking voltage between case earth and component terminal)
C -9v/-2.7v
B 2.3v/4.9v
E 0v/0v

73's de Edd
But first, would you interchange the U302 and U304 parts positions, as it is, the
U302 is having the least power demanding circuit application, of those two..

Hi 73's de Edd, Just checking as these two are different components but ill trust your superior knowledge if your numbers are correct.

Cheers

Key

 

[shrtrnd]

Since you hope to turn this into a hobby, I suggest buying a can of 'freeze spray' from an electronics supply store.
When you have a problem like this, spray the suspected component (usually a semiconductor) to cool it. When you
hit the right one, the unit will function again for a few seconds until the part heats-up again. That's how you quickly
ID a defective semiconductor when it fails due to heating.

 

[Keyhall]

Hi shrtrnd, never thought of that, I think I have some knocking around in the garage will take a look. Am i right is saying that some of these components have built in temp monitoring and shutdown protection or would another part of the pcb detect the temp?

 

[Keyhall]

Update, I have just done a diode test on U304 accross the B C E (whilst still on the board), and get the same readings both before and after the sound cuts off so think this rules out this component as the issue..

 

[shrtrnd]

Very few, very expensive electronic components have internal shut-down capability. What this sounds like to me,
is a semiconductor failing when it heats-up. It's pretty common when one is on the way out. The method I suggested
is the quickest way to verify that, and which component is failing. Good luck with troubleshooting.

 

[Keyhall]

Thanks, I found my can of freeze spray and tried what you suggested on various components in the area but no obvious positive results as yet, will keep trying.

 

[Keyhall]

Right might have found something...just removed capacitors C312, C311, C313 & C314 and tested using a little gizmo I purchased a while back.

Three read around the 22uf range with an ESR of 30 to 52 ohms
but the other reads 19.86uf and an ESR of .14K which I believe to be 140 ohms which i think is outside the tolerance and the ESR is way off.
could this be a sign of a worn/on its way out capacitor?

please say it is

 

[shrtrnd]

It'll make you happy to say yes, but it may or may not be your problem.
Sure, replace the suspect cap and see if you've isolated the fault yet or not.
If this was me, and the fact that the caps are cheap, would make me replace all three caps.
They all came out of the factory at the same time, and if one is 'leaking', odds are the others may not be far behind.
It's not unusual for caps to go bad by themselves, but they're often stressed to the point of impending failure by
the current through the cap from another failing component further upstream.
Replace the caps, and 'freeze' spray the nearby components feeding those caps.
I'm keeping my fingers crossed for you.

 

[73's de Edd]

Sir Keyhall. . . . . . .


I was not absolutely, positively,112% certain of any casing configuration variant between an LM337T and its AT version .
I was expecting both to be leaded, thru hole mount TO-220 cases.

Possibly you are experiencing /seeing a difference by virtue of one possibly being a three lead through hole configuration, while the other might be a bent tab flat mount version . Is that your viewed situation ?
In either case, I see very little power demand being placed upon them.
The REAL power regulation is relying upon the Q301 and Q302 main power transistors with an assist from their associative U301 and U302 units.

On your electrolytic capacitor situation, with all of them being 22 ufd, one might expect normal aluminum units to be exhibiting anywhere from 7 to 30 ohms of ESR, while the value of "Pwemium bwewed" low ESR units might be down in the 1 through 5 ohm ESR values..
You definitely have a possibility on that 140 ohm unit, but you neglected to say what its particular location and its assigned number is, on the schematic
But it seems like that the 15 positive and negative regulators are using a handful of 22 ufd marked units throughout their circuitry .

Looking at the schema the MINUS 15 VDC circuitry, relies upon
C312 and C314 AND if input filter C310 has an elevated ESR, it will make U304 run hotter.
BUT I normally expect MINIMAL power demands on the negative branch of this type of minor dual supply.
Check ALL across the WHOLE schematic and just find out how many power derived connections are being made into either a marked -15V or its like marked PP213 circle.

I had never even heard of your brand of amplifier, looks like it might be of 1990 vintage, also there is a possibility that these 22ufd electrolytics mentioned might be surface mounted mini cans and of the type which I am showing just below.

If your units capacitors are of being of this type, there was a total rash of failures of this type back in that time frame.
A good amount were "counterfeit" units that had utilized an improper formulation of their internal electrolyte.
WHILE one popular manufacturer just happened to have erred in his amount of electrolyte put in them, and doing an OVER fill. Those units later seeped out CORROSIVE !, pcb blackening electrolyte, or if used in a fast switching power supply application that caused them to overheat, their bottom seals ruptured.

On this power supply of yours, the principal supplies are its plus and minus 26VDC supplies, with your smaller plus and minus 15VDC supplies branching off of those main plus and minus 26VDC supplies. .
The major power demand is pulled by your audio output stages, while THESE plus and minus 15VDC supplies,would be providing a minor supply level.



73's de Edd

 

[Krapoman]

Hi everyone,
This trick with the freezer is great.

well, i'm a bit confused. I wonder how the machine can work correctly when it's cold, thought the voltage reading on the U304 out terminal is low since the beginning..but i'm sure there an easy explanation.

Check ALL across the WHOLE schematic and just find out how many power derived connections are being made into either a marked -15V or its like marked PP213 circle.

if i understand anything, the +15/-15 supplies the lower stages throught Transistors in common collector. The output voltage there should be 15-0,6=14,4v
You can see these circuitries at G6 and B8 on the map.

and in the Headphone stage, The +15/-15 is wired throught reversed inline diodes..I don't understand what's the use.

But that's not the point.

If i'm right, you should read 1,4v at the base of Q307, not less..

here's what was my calculation :
-15/+15 throught a resistive bridge of 120k + 100k = 1,4v at the middle point ( pp 79 ).

but according to the readings, i m totally wrong

U304 before after
in -25v -25v
out -8v -3v
adj -7v -3v

Q307 before/after sound cutt off
C -9v/-2.7v
B 2.3v/4.9v
E 0v/0v

what about the two Diodes D304 and D305 ?
I don't understand how C313, D304 and R326 interact with Q307, but it looks like some kind of regulation.

 

[Keyhall]

thanks for all the info..I have decided that since this a high end amp and intend to sell on once fixed I'm going to replace all the capacitors on the board for top quality rubycon items from Farnell.
once they have arrived and I've got them fitted I'll report back with an update. hopefully a positive one.
thanks for now.

 

[Keyhall]

ps. on the location of the possible faulty capacitor I made a stupid mistake of not noting the location on the board as I removed all four in one go before testing.
the capacitor are of the normal leaded aluminium can type and all four transistors (LM337T) etc are as you have displayed in the image above so if it is safe to swap u302 and u304 around I will give that a try if changing all the capacitors around doesn't solve the problem.

 

[73's de Edd]

Sir Keyhall . . . . . .


It's good to know that your unit is using standard leaded capacitors instead of that surface mount type, which I showed you .( Those were hard to replace / work with. )
First of all let's get you COMPLETELY straight on those LM337 " transistors " , as they are actually being a complex integrated circuit that, just happen to then use only 3 external leads.
GOOGLE yourself LM337 and get a data sheet and on one of its pages you will see exactly how complex the innards of that unit is being.
Also, you are referring to 337 units being used, when in actuality, there are two DIFFERENT types being used.
You need to watch their second digit VERY closely as the 1 is designating a positive type of voltage regulator while the 3 is designating a negative type of voltage regulator.

WORDS OF WISDOM:

You should have initially tested your capacitors left in circuit with your ESR tester.
Why is that ?
BECAUSE:
Soldering iron heat, that will be carried by the wire leads into the internals of the capacitor can cause a miraculous , yet , VERY short time lived curative / healing effect upon the capacitors ESR value.
On the two 22ufd capacitors associated with your U304 circuitry, the most critical would be C312 with the next most critical being C 314 .


73's de Edd

 

[Keyhall]

Hi guys, sorry for the long delay but I have an update, I replaced the 4 caps that I had originally removed and it has fixed the problem. I've had the amp playing for hours since and have had no issues at all. Looks like the high ESR cap was the culprit.

Just one last very newbie question if i may,

I cant get my head around how to work out what the voltage drop should be across a resistor.
If I have 4 volts going into a 22ohm resistor what should the voltage coming out be? I know Ohms law requires the current required but I have no idea where to get this figure from?

Thanks for all your help

 

[73's de Edd]

Gonna sell that unit for BIG G G G G G G BUX now . . . .Huuunnnnh


4 volts going into a 22 ohm resistor what should the voltage coming out be?




It will still be 4 Volts UNTIL you put a load /resistance on its output to make a COMPLETE loop feeding back to the power source them it divides ratiometrically.
If you took that first 22 ohm resistor and series added another 22 ohm and the second resistor completes a power loop. You would then have a 2 volt reading where you initially had the 4 V unloaded conditions reading.
Monitoring the current in the loop, one would find 90 ma.


73's de Edd

 

[Keyhall]

Thats the plan.

The reason i ask the question is that I would prefer to test resistors whilst still on the board using the voltages rather than across the resistor using an ohmmeter as that can give inaccurate readings.
I guess its not that simple and will need to understand schematic diagrams a little better in order to do this.
Cheers