Lm1875 - need some help..

[R.G.]

The shematic I used was the one labeled "Typical Applications" on page 1 of the PDF. It´s the split power supply schematic.

Sorry, I must be extra thick tonight. Which PDF? Which typical applications? The ones on the LM1875 datasheet?

[Threefish]

There is a split supply schem on the first page of the National Semi app note. That's the one you're referring to?
If you have the time/patience, do a tracing of your layout and post it. There may be something simple that you've reversed or soldered to the wrong bit.

In my recent post about my LM1875 amp (different problem - solved!) I've attached the schem that I used, laid out slightly differently (and with a couple of slightly different and unnecessary components) to that app note one -
http://www.diystompboxes.com/smfforum/index.php?topic=61889.0
Compare yours to it. I think the app note schem is badly drawn. It's a bit screwy where the -ve supply comes in. For my brain anyway.

Good luck. Keep at it.

Reid.

[bamera]

I bought a new IC, carefully put together another circuit and tested it. Same problem as before, nice sound for about 10 sec then shutdown.

This problem must something to do with the PSU caps charging.  As I was taking down the pin voltages I accidentally shorted some pins with my DMM, there was a loud speaker thump and flash, This short kick started the amp again with fuzzy sound slowly fading in until max volume and Cristal clear.   After a few seconds of nice sounds the amp shutdown again. 

Hook up your DC meter across your loudspeaker, set to a voltage range of 20V or more. Turn on the amp and see if there is DC across the speaker, and more importantly, does it slowly increase across 10 seconds.

Voltage across speaker: -1.2 mv stable until shutdown

Nail down the exact schematic, the exact pin voltages, and in this case the pin voltages with time.

Time is about 10 sec from startup to shutdown.

Pin 1 :   1.8mv slow increase to 2.2mv at shutown. ( 1.06v with amp in shutdown)

Pin 2 :   0.6mv slow increase to 1.2mv at shutdown. ( 1.6v with amp in shutdown)

Pin 3 :   -22.2v quick decrease to -0.72 at shutdown. ( 0.7v with amp at shutdown)

Pin 4 :   -1.0mv stable until shutdown.

Pin 5 :   24.2v quick increase to 48.0v at shutdown. (48.0v with amp in shutdown)


I hope this makes sense,

Daniel C

[bamera]

The shematic I used was the one labeled "Typical Applications" on page 1 of the PDF. It´s the split power supply schematic.

Sorry, I must be extra thick tonight. Which PDF? Which typical applications? The ones on the LM1875 datasheet?

Yup, I used the one on the LM1875 datasheet. This is the one i used. http://www.national.com/mpf/LM/LM1875.html

[bamera]

There is a split supply schem on the first page of the National Semi app note. That's the one you're referring to?
If you have the time/patience, do a tracing of your layout and post it. There may be something simple that you've reversed or soldered to the wrong bit.

In my recent post about my LM1875 amp (different problem - solved!) I've attached the schem that I used, laid out slightly differently (and with a couple of slightly different and unnecessary components) to that app note one -
http://www.diystompboxes.com/smfforum/index.php?topic=61889.0
Compare yours to it. I think the app note schem is badly drawn. It's a bit screwy where the -ve supply comes in. For my brain anyway.

Good luck. Keep at it.

Reid.

Reid,

I read your post along with some others while looking for an answer to my problem, I´ll see if I can put up a tracing.  The -ve on the schematic left me thinking too!!!!

[R.G.]

We'll wait for the layout, but it's obvious from the voltages that your amp is suddenly pulling the speaker to V-, but the V- can't support the speaker load and instead of the speaker voltage moving, the amp is pulled to the bottom side of the power supply.

This strongly suggests a wiring error. We need the wiring/tracing to determine more.

[bamera]

R.G - I´m glad something made sense with those values 

I had the C5 0.22uF incorrectly placed with a 22uF eletro, I switched it to the correct value (0.22uF) and tested the amp. No difference that I could tell, a few seconds of sound and then nothing. The problem continues.
I hope this doesn´t effect the pin voltages I posted.

Below is an almost exact tracing of my circuit. The black wires outside the PCB are the grounds which are all connected to a wire segment that joins them all to one point on the PCB itself.
With exception of the -V input.



Can´t seem to insert the image

[bamera]

Here´s the layout..


http://img219.imageshack.us/img219/5245/lm1875layoutwn9.jpg

[R.G.]

It's not clear from your layout where the wires go to the power supply. Which wires go to the power supply center? Which ones go to +24V, which go to -24V, and which go to ground? Does the input jack "ground" connect to the power supply ground? It doesn't appear to. And I can't tell about where the rest of the wires go.

[bamera]

It's not clear from your layout where the wires go to the power supply. Which wires go to the power supply center? Which ones go to +24V, which go to -24V, and which go to ground? Does the input jack "ground" connect to the power supply ground? It doesn't appear to. And I can't tell about where the rest of the wires go.

Sill learning as i go



Here goes a more detailed picture R.G, I tried to cover all the points you mentioned but to be clear, in the new picture all of the circuit ground wires meet at the top left corner of the PCB, the ground ends there and is not connected to anything. (This may be unclear in the picture but dragging those wires looked too messy)

I do have two questions though.



-The input jack ground does not connect to the power supply ground, actually the input ground is my star grounding point. Should the power supply gound be connected here?

-By the power supply ground do you mean the center tap  0 V?


Thanks for sticking with me on this

Daniel C


If this link doesn´t work try the bottom one


[img=http://img66.imageshack.us/img66/3162/lm1875layoutpo0.th.jpg]

[R.G.]

-The input jack ground does not connect to the power supply ground, actually the input ground is my star grounding point. Should the power supply gound be connected here?
-By the power supply ground do you mean the center tap  0 V?

And this is probably your problem. Yes, the input ground must connect to the power supply center tap ground. The center tap ground is where all of the electrical currents eventually wind up. By not connecting the input and/or star grounds, you build up voltage on the decoupling caps and that's what turns it off after a few seconds.

Try connecting your input ground to your star ground with a wire, and run another wire over the power supply center tap 0V. The speaker wire should go directly to the power supply center tap ground.

[PerroGrande]

Yep -- that's gotta be it.  The center-tap in a split supply ends up being the ground reference point for both "sides" of the supply. 

I'm a bit curious as to why one of the three different types of amp chips you used worked at all...

[bamera]

IT WORKED!!!!  Thanks R.G for pestering me into giving you all the minute details    I wouldn´t have worked if you hadn´t!!!
You are a gentleman, master of all things electric and a connoisseur of fine cheese and wine.

Yep -- that's gotta be it.  The center-tap in a split supply ends up being the ground reference point for both "sides" of the supply. 

I'm a bit curious as to why one of the three different types of amp chips you used worked at all...

Thank you too PerroGrande for taking an interest and making me actually think a little

Yes, my TDA2050 works very well without the zero volt connection, but since I´ve only worked with the TDA20XX series  (All worked fine). I just tried connecting the zero volt to the 2050 a few minutes ago and the only difference was louder radio interference. I guess it was time to learn someting new.

Just to remind you of my first post, my LM3886 presented a similar problem to this one only that the shutdown was not immediate, the sound would continue although with major distortion and lower volume. Eventually it would shut down but at louder volumes it would continue without shutdown but functioning poorly.

Thanks again for all the help guys, I really appreciate it.

Daniel C

[R.G.]

I'm a bit curious as to why one of the three different types of amp chips you used worked at all...

I've run into something similar. With the power supply ground disconnected, the signal ground stays at a voltage that is determined by the amount of current which is drained from or forced into it. If these are equal at some voltage, the 'ground' settles to that voltage and stays there are equilibrium. If either the up or down currents are mismatched, 'ground' walks the direction it's being forced until something quits.

One of the earlier setups hit a rough equilibrium somewhere within the power supply range. This is much like picking resistor values for a bias voltage, for which, see GEO.

[Threefish]

Glad to see you're up and running, and that I could help a little (if nothing else, prod you about your layout).
I'm still really a noob at this. I don't get the satisfaction of helping very often.   

Best,

Reid.