Solid State Guitar Amp Forum | DIY Guitar Amplifiers

Solid State Amplifiers => Amplifier Discussion => Topic started by: kin0 on May 11, 2011, 11:09:07 AM

Title: debugging a small amp
Post by: kin0 on May 11, 2011, 11:09:07 AM
Hay, today I finished building my first ss amp. It's a 8w chipamp kit for the power amp and the barber ltd (silver fox clone) for preamp. The power amp worked great and made sounds but after I added the preamp I can just hear white noise (then I turn on the drive channel i hear more gain but still no sound). What places to check and how to debug?
Here is the topic with my project planning (in the end I put a barber ltd instead of the shredmaster and used only 1 drive pedal instead of 2)

http://www.ssguitar.com/index.php?topic=1834.msg12006#msg12006 (http://www.ssguitar.com/index.php?topic=1834.msg12006#msg12006)
Title: Re: debugging a small amp
Post by: tonyharker on May 11, 2011, 11:39:34 AM
Sounds like the preamp is at fault.  Did you build it yourself from a kit or was it ready made?  Do you have the circuit for it? 
Title: Re: debugging a small amp
Post by: kin0 on May 11, 2011, 01:04:09 PM
I made the pcb myself. Here it is:
I can't find form where I downloaded it.
(http://up352.siz.co.il/up2/mvjyyahzftwm.bmp) (http://www.siz.co.il/)
http://www.siz.co.il/my/mvjyyahzftwm.bmp (http://www.siz.co.il/my/mvjyyahzftwm.bmp)
oh and this is the power amp kit.
http://www.vellemanusa.com/us/enu/product/view/?id=350529 (http://www.vellemanusa.com/us/enu/product/view/?id=350529)

Oh, im so stupid. I forgot in insert the ic in the preamp. But it still don't explains why it haven't made any sound in the clean channel and why it made so much noise.
Title: Re: debugging a small amp
Post by: joecool85 on May 11, 2011, 01:39:45 PM
Wait...no IC would mean no sound.  Have you tried it now that you have the IC in?
Title: Re: debugging a small amp
Post by: kin0 on May 12, 2011, 12:04:23 AM
the problem it haven't worked even then I bypassed the overdrive. By the way I used a dpdt switch. 1 Pole is switching out put and 1 is switching the led (the inputs are joined together). Is this a good switching?
Title: Re: debugging a small amp
Post by: kin0 on May 16, 2011, 08:30:54 AM
okay now I got the power amp working. A power wire got loose. I also put in the IC. the clean (staright to the power amp) works but the dist channel still just makes noise. Also I put a pot before the power amp (as it's said here in page 8:http://www.vellemanusa.com/downloads/0/illustrated/illustrated_assembly_manual_k4001.pdf (http://www.vellemanusa.com/downloads/0/illustrated/illustrated_assembly_manual_k4001.pdf) -except I put it between a dpdt so it would work only with the clean) but it does nothing the volume stays the same.
Oh and there are some buzz then the volume is up. In my guitar amp there is no buzz at the same volume.
Title: Re: debugging a small amp
Post by: joecool85 on May 16, 2011, 11:01:11 AM
okay now I got the power amp working. A power wire got loose. I also put in the IC. the clean (staright to the power amp) works but the dist channel still just makes noise. Also I put a pot before the power amp (as it's said here in page 8:http://www.vellemanusa.com/downloads/0/illustrated/illustrated_assembly_manual_k4001.pdf (http://www.vellemanusa.com/downloads/0/illustrated/illustrated_assembly_manual_k4001.pdf) -except I put it between a dpdt so it would work only with the clean) but it does nothing the volume stays the same.
Oh and there are some buzz then the volume is up. In my guitar amp there is no buzz at the same volume.

I would remove the pot and dpdt switch, sounds like they may be suspect in this case.  Let us know if that gets it working on the clean and OD sections.
Title: Re: debugging a small amp
Post by: kin0 on May 18, 2011, 08:58:18 AM
KK
Title: Re: debugging a small amp
Post by: kin0 on May 21, 2011, 12:56:54 PM
I tried to take out the pot the the switch. The clean works (straight to preamp) but the drive channel (through the pedal) don't work-just makes lots of noise.
Title: Re: debugging a small amp
Post by: joecool85 on May 23, 2011, 08:28:11 AM
When you say "drive channel (through the pedal)", do you mean you are running a pedal in front of the clean circuit?  Or is it it's own preamp entirely?

I ask this because I don't see two channels in the preamp schematic you showed us.
Title: Re: debugging a small amp
Post by: kin0 on May 27, 2011, 08:13:21 AM
Clean I mean directly to the power amp and by drive channel i mean when i put the stompbox circuit i posted before the power amp
Title: Re: debugging a small amp
Post by: phatt on May 27, 2011, 10:32:35 AM
Hi Kin0,
           R11 does not look right and the bas/treb looks rather odd also?
i'd be checking that the schematic actually Valid and known to work.

Also your board has a earth traversing all the way round the outer edge.
that is a loop and can cause issues,, but should not stop it from working.

R11 normally goes through a Capacitor to ground not bias voltage.
Check on other circuit that are known to work to give you some ideas.
Personally I would not use this circuit as it would be prone to problems,,, but I've not built that exact circuit so who knows maybe it would.

I found tone circuits are always *After distortion circuits* and I've had far more success by implimenting tone stuff *Before Dist* my 2 cents :tu:
Phil.

Title: Re: debugging a small amp
Post by: kin0 on May 27, 2011, 01:53:18 PM
I don't think that there is a problem with the circuit. This is a barber LTD clone and it uses the exact schematic that is given in Barber site.
Title: Re: debugging a small amp
Post by: joecool85 on May 27, 2011, 02:01:07 PM
Have you tried running the OD circuit into a known working amp?
Title: Re: debugging a small amp
Post by: J M Fahey on May 27, 2011, 05:02:39 PM
Quote
This is a barber LTD clone and it uses the exact schematic that is given in Barber site.
Fine, congratulations, re-check it anyway.
Even after thousands of amplifiers, each new one I finish building goes through the series lamp first, no exceptions.
Most times, say  95% , it comes alive with no hitches; yet a few here and there do not.
There's always the very minor problem which rises its head at intervals.
Oh well.
That's why even monster robotized factories such as Fender, Marshall, etc, regularly pull some amps from the assembly line for special checking, and most serious ones check *all* (even if only for one minute) , at least by strumming a few chords through them.
I have somewhere the famous picture of a *very* plain looking Chinese girl, clad in blue worker's overalls, with spectacles and ponytail, playing a very evil looking Death Metal guitar through Bugera heads prior to packing and shipping.
So, in a nutshell, building something apparently "by the book" is no 100% guarantee that it will work as expected or not need any testing.
Title: Re: debugging a small amp
Post by: joecool85 on May 31, 2011, 11:29:10 AM
I agree with Juan.  No matter how good you are, mistakes happen.  Best to go through it and check for any issues.
Title: Re: debugging a small amp
Post by: kin0 on June 25, 2011, 10:25:23 AM
Okay, I don't really know how this stompbox should be but I decided to check if the schematic the the PCB are the same (after that I'll follow this guide:http://www.geofex.com/fxdebug/newfx.htm (http://www.geofex.com/fxdebug/newfx.htm))

I have some question:
1. What is the purpose of the red section?
2. Where the 9V (blue) should connect to?
3. All the VB points connect together but what the hell is VB?

(http://up352.siz.co.il/up1/ygo0ehdtjnlm.bmp) (http://www.siz.co.il/)
Title: Re: debugging a small amp
Post by: Alexius II on June 25, 2011, 11:35:31 AM
I think I can provide an answer:

1. the red section is "power supply" (led power, filtering, provides a reference voltage)
2. 9V (blue) should probably connect to the + side of that 100uF capacitor at the power supply
3. VB is just a name for a reference voltage; in this case I believe VB is at 4.5V (or half the supply voltage)

Hope this helps. Please someone correct me, if I'm wrong  ::)
Title: Re: debugging a small amp
Post by: kin0 on June 25, 2011, 11:38:35 AM
Okay, The 9v really connects there and the part after the 100uf cap is to get 4.5volt but why there is a grounded diode and a resistor between the 9V source and the 9V+?

I checked the parts values and discovered some things.
Instead of 13.3k resistor I got 12k, instead of 27n cap I got 33n, instead of 20n cap I got 22n. Is getting these exact values is very important or is it okay to leave it just as it is?

Also I got mixed with the group of 4 resistors (D1 to D4). In the schematic the upper couple are turning left and the lower couple are turning right, in the pcb the upper couple turning right and the lower turning left. I did it as in the pcb. Is this important that I'll change them to be as in the schematic or it will work just fine if they'll stay as the pcb says.

I need to check the connection on the pcb but first I need to change a battery in my multimeter-I'll check the solder joints on the same time
Title: Re: debugging a small amp
Post by: phatt on June 26, 2011, 06:43:23 AM
Hi kino,,
           D5 is simply to save the circuit from reversed polarity,, battery connected backwards etc.

22nF is close enough to 20nF. as with most of these values it's not super critical.
If you must have 13.3k then 10k + 3k3,, again hardly worth the fuss as 12 k is likely close enough.

As mentioned VB is the offset voltage, or VB (Voltage bias)
to get these devices to work from a single supply you need to lift the input halfway,, in this case 4.5 VDC is the bias point.

Then of course they need to be decoupled via a capacitor for them to work.
R13 is not needed on Battery but if you run from a plug pak with other devices connected to same it can help filtering issues.
Phil.
Title: Re: debugging a small amp
Post by: phatt on June 26, 2011, 06:44:49 AM

Oh forgot the other diodes.
                                      Two go forward two go backwards otherwise it won't work as intended.
Phil.
Title: Re: debugging a small amp
Post by: kin0 on June 26, 2011, 12:26:20 PM
my question was is it important which pair goes forward and which pair backward or not?
And thanks for the other answers
Title: Re: debugging a small amp
Post by: Alexius II on June 26, 2011, 12:57:18 PM
my question was is it important which pair goes forward and which pair backward or not?
And thanks for the other answers
If all diodes are of the same type, than no, it is not important.  :tu:
Title: Re: debugging a small amp
Post by: kin0 on June 26, 2011, 01:05:26 PM
oh crap now I have to look for another problem. Cool smili by the way. ;)
Title: Re: debugging a small amp
Post by: kin0 on July 06, 2011, 11:53:12 AM
I checked if everything on the pcb is connected well. There were some places that have to be connected (not a problem) but there are some places that are connected to the ground while the shouldn't be. I can't find the exact place that they are connected (to know there to disconnect). Here is there I have the problems.

(http://up352.siz.co.il/up1/juiy02zjmxmz.bmp) (http://www.siz.co.il/)
Title: Re: debugging a small amp
Post by: J M Fahey on July 06, 2011, 12:56:36 PM
Label them with a letter, for clarity.
What do the red lines mean?
Both ends are connected?
Either end is ground?
Title: Re: debugging a small amp
Post by: kin0 on July 06, 2011, 04:29:01 PM
the red lines mean that i put the multi meter in the both ends of the line and it made a sound (means that these two points are connected to each other somhow). In several of these point I can see clearly where they are connected but in others I just can't see anything wrong. How can I find what to fix in those that I can detect any visual connections.
Title: Re: debugging a small amp
Post by: phatt on July 06, 2011, 06:24:24 PM
Kin0,
       You will have better results if you learn to Work in Nodes,, not just what the layout looks like.
Example;
            The very first input has the Input, R1 and C1.
All 3 connect to one point/Node.
ONLY Those 3 components connect to that place on the circuit board.
If you number each NODE on the schematic it will make the whole layout more clear.
Have fun,, Phil.
Title: Re: debugging a small amp
Post by: J M Fahey on July 07, 2011, 01:40:52 AM
Please re-post labelling the "red lines" to be able to give you a useful answer.
As in: "a", "b", "c", etc.
Title: Re: debugging a small amp
Post by: kin0 on July 07, 2011, 11:30:10 AM
I don't understand what do you mean  :o
Title: Re: debugging a small amp
Post by: J M Fahey on July 08, 2011, 02:53:09 AM
You are showing 10 red lines.
It's very tiring trying to find a unique description for each of them, such as "the short squiggly one which is to the right but not *that* much to the right of R1, I mean the end of R1 that joins C2, not the one which touches R32 ..... " and so on.
When you actually find *which* red line I am talking about, your liquefied brain is flowing through your nose.
Not many neurones alive, after that, to understand my suggestions.
Label each of them with a letter.
After all, there *must* be some reason for designers to label resistors as: R1, R2, ..... R99 ; capacitors C1, C2, .... etc.
And as Phatt said, understanding nodes; meaning node is what connects many points together, is great to understand schematics (and PCB Layouts).
In fact, schematic design software exports long lists of nodes, which describe connections within a PCB, which will later be made real through copper tracks..
And don't worry about writing
Quote
I don't understand what do you mean
, we are *all* learning here.  :tu:
Title: Re: debugging a small amp
Post by: tonyharker on July 08, 2011, 04:48:35 AM
You cant check a circuit like this using a multimeter in this way.  The circuit components will give you false indications. Especially if they are active.  Unless you are checking the board with no components on.
Title: Re: debugging a small amp
Post by: kin0 on July 08, 2011, 09:01:39 AM
I can't unsolder all the components O.O. What I did is to check the connection between each leg, I don't think that a components will do any different.

Here is the new layout. The brown point are all the ground. F is a line which is connected to the ground in every point of it (got 3 components connected on it)

(http://up352.siz.co.il/up1/gnigqmmmg4fx.bmp) (http://www.siz.co.il/)
Title: Re: debugging a small amp
Post by: phatt on July 09, 2011, 10:00:41 AM
Hi Kino,
        Lets start again shall we?
We have established It makes a noise but passes no signal.

Check you have DCV in the places I've marked pins 8 and 4 are the DC power

conections, so check this first.
Next,,,Pins 3 and 5 need some kind of DC offset voltage so check it.

then we move on to signal path.

something does bother me bout this circuit:(

There is no De-coupling cap after u1 which does not seem correct.
The circled *ground I've marked* at the tone circuit maybe wrong, others here may

know more but to my mind it needs to return to the bias node otherwise it may not

work correctly.
Phil.
Title: Re: debugging a small amp
Post by: kin0 on July 14, 2011, 10:19:50 AM
Okay I fixed every problem I had except the false connection between E and ground. (Phatt I'll check these things if it won't work after this one).
Is there any connection there to the ground?
Title: Re: debugging a small amp
Post by: J M Fahey on July 14, 2011, 10:33:03 AM
When you find an input grounded (when it shouldn't) the *very* common cause is that the soldering heat melts the plastic insulation and lets the hot wire touch the screen.
It happens to me even today.
Title: Re: debugging a small amp
Post by: kin0 on July 14, 2011, 12:56:01 PM
but there is no plastic insulation to melt O.O
Title: Re: debugging a small amp
Post by: J M Fahey on July 14, 2011, 07:26:14 PM
Don't you use screened wire for signal input?
Please post a picture so we know better what we are talking about.
We can all see the *schematic*, which is a bunch of symbols, but nobody knows *how* it was built or wired.
Title: Re: debugging a small amp
Post by: kin0 on July 16, 2011, 05:51:56 AM
But the input don't touch the ground. Line E which connects the presence trimpot leg to the C3. This line somehow touches the ground and I need to disconnect it from the ground (I can't see any visual connection).
Title: Re: debugging a small amp
Post by: phatt on July 16, 2011, 07:15:40 AM
Then remove the offending parts, remelt any solder and suck any blobs away until you have a clean, clearly visible track.

Then redo those components.

Ed, without close up pics of the board and offending components this is going to take a long time for us to help.
Phil.
Title: Re: debugging a small amp
Post by: kin0 on July 20, 2011, 09:21:11 AM
I fixed the ground problem and now I have some problems with the electricity.

The red 9v+ (the 9v input to the circuit) get 8-9v, but the blue 9v and the VB get only 0.2v. How can I fix it?

http://imageshack.us/photo/my-images/24/unledjmr.png/ (http://imageshack.us/photo/my-images/24/unledjmr.png/)
(http://imageshack.us/photo/my-images/24/unledjmr.png/)
Title: Re: debugging a small amp
Post by: phatt on July 20, 2011, 10:03:05 AM
If 9VDC at Node (D5,R13,R16) then you should have close to 9VDC at pin 8 of the IC.
If not check polarity of C11, check C12 while you are at it.
Phil.
Title: Re: debugging a small amp
Post by: kin0 on July 31, 2011, 10:09:45 AM
The cap polarity is okay. And the node get 8.25v (idk why not 9) but the connection of the electricity to pin 8 of the chip is very strange. Instead of going straight from the 9V to the pin (red) it goes from R13 to the pin (blue). Is this the problem? And what can be the other reason for the VB not getting the electricity?
Title: Re: debugging a small amp
Post by: phatt on July 31, 2011, 10:33:48 AM
Blue is correct. and there is always a small voltage drop across the 91 Ohm resistor.

You are obviously struggling with this,, why not take some pics of the actual board if you can?
Preferably both sides.

Ed,, check that R14 and 15 are in fact 10k (usually Brown Black Orange)
You may have got the 470k R in the wrong place and that would certainly stop it working :'(
Phil.
Title: Re: debugging a small amp
Post by: J M Fahey on July 31, 2011, 11:29:53 AM
Agree, you have some PSU ripple filtering courtesy of R13 and C11.
The price you pay is a slight voltage loss.
Nothing on Earth is free.
Well, maybe true Love  <3) , but not much else.
Also agree on debugging: cut the track that joins the node R14/R15 to C12/R2 and measure voltage across R14 and R15.
Post it.
Title: Re: debugging a small amp
Post by: kin0 on July 31, 2011, 03:57:01 PM
How can I cut the track?
Title: Re: debugging a small amp
Post by: J M Fahey on August 01, 2011, 12:56:39 AM
Make 2 parallel cuts across it with a sharp X_Acto type knife, the so called box-cutters sold for $1 everywhere.
(http://upload.wikimedia.org/wikipedia/commons/thumb/c/cf/Box-cutter.jpg/220px-Box-cutter.jpg)
They must be parallel and separate by 1 mm; then you can lift the thin center slice either by cutting somewhat sideways, or putting tour hot soldering iron tip on it and pulling it away, because the hot tip weakens (a lot) the glue that holds it.
You measure what you have to, and then you lightly scratch both sides of the slit and re-join them with a drop of solder.
Title: Re: debugging a small amp
Post by: kin0 on August 08, 2011, 08:55:44 AM
gets only 0.25V
Title: Re: debugging a small amp
Post by: J M Fahey on August 08, 2011, 12:08:10 PM
Quote
measure voltage across R14 and R15
I mean what voltage you get across (from end to end) R14 and what you get across R15. 2 values in total.
Even better, post the value to ground (black probe grounded at the battery, red probe measures) on each of R14 and R15 legs, that makes 4 values in total.
Sorry if I wasn't clear.
PS: I guess you cut the approppriate track so the node that joins them is connected to nothing else.
Title: Re: debugging a small amp
Post by: Casey4s on August 08, 2011, 02:13:05 PM
Great thread  :tu:
Title: Re: debugging a small amp
Post by: kin0 on August 08, 2011, 04:32:43 PM
Yeah I cut the connection.

The node of the R14/R15 is 0.24
the other part of the R15 is 0
and the other part of the R14 0.9
Title: Re: debugging a small amp
Post by: J M Fahey on August 08, 2011, 05:17:34 PM
Ok.
Quote
The node of the R14/R15 is 0.24
Now you are talking professionally :tu:
Even if it looks slightly longer or more complex, it ends being the shortest and clearest way.
So R14 itself is getting *only* 0.9V through R13, instead of the almost 9V we expect there.
Why?. Not normal.
Looking at the PCB rather than the schematic, I see a node R13/R14/C11/IC1Pin8.
There's *something" there pulling voltage down; it may be, in no particular order:
1) C11 inverted (upside down)
2) R14/R15 color bands misread, they are 1/10th the expected value (not very likely, but sh*t happens)
3) R13 misread, 10X the expected value (same considerations)
4) IC1 bad/shorted/upside_down, "eating" tons more than it should.
I would pull it (carefully, pampering the solder pads, use solder wick if available or a very good solder sucker)
5) bad connection in the +9V rail: bad soldering, connector, cracked track, you name it.
Ther is no magic nor gremlins, but sometimes an apparently simple problem can drive you crazy.
See that for this very simple one I am suggesting 5 or 6 possibilities, that's why when somebody posts "xxx circuit does not work, what may be happening?" and offers no other clues,  it's impossible to provide a useful answer, the asker must do his part of the job too.
OK, less talk (I), you go measure.
Now you are getting the repair/analyze logic. :tu:
Title: Re: debugging a small amp
Post by: kin0 on August 09, 2011, 12:41:15 PM
The opamp was inversted (Sockets f!!* YEAH  8) ).
Now the node of R14/R15 is 1.7V
the other side of R14 is 9V
and the other side of the R15 is 0V

The opamp (pin 8) is now getting 9V too.

Is it okay now?
Title: Re: debugging a small amp
Post by: joecool85 on August 09, 2011, 01:01:43 PM
Is it okay now?

You tell us, does it work?
Title: Re: debugging a small amp
Post by: kin0 on August 09, 2011, 01:08:20 PM
I have to connect back the R14/R15 node to the to C12/C2 so I prefer to know first if the voltages should be like that. BTW all the other 4 things are right
Title: Re: debugging a small amp
Post by: J M Fahey on August 09, 2011, 04:27:48 PM
*If* N(ode)-R14/C11/R13 is 9V, on N-R15/R14/C12 you *must* have half that, since both R14 and R15 are same value, 10K.(check it)
You do not  :(
C12 might be inverted/upside down (make sure negative goes to ground, positive to Vb)
IC1 might be bad after accidental inversion.
Measure Vb , first without IC , then without C12 if necessary.

One of the general-purpose repair systems is:
 Ohm's Law rules; and lets you calculate what you *should* find at different points in the circuit.
If you don't, imagine what might shift those values and check.
As a non-electronic example:
your car engine does not start.
* do you have a spark?
* does gasoline reach the cylinder? if not, why?
* Does it reach the carburetor?
* Does it reach the gas filter?
* is tha gas pump working ?
* any problem with the hoses and conduits ?
* do you have gasoline ?
it's the same.

Title: Re: debugging a small amp
Post by: joecool85 on August 10, 2011, 06:41:54 AM
* Does it reach the carburetor?

You guys still running carbs down there?  We haven't had a mass produced vehicle with a carburettor since 1985 or so.
Title: Re: debugging a small amp
Post by: J M Fahey on August 10, 2011, 08:02:11 AM
Not really.
It was just an example of what troubleshooting means, not only in Electronics but anywhere else.
H*ck !!, my Father was an Old style Country Doctor and he used basically the same mindset to "troubleshoot" PEOPLE !!   ;D

Back to cars, you have *any* car in the world available here, thanks to "Globalization".
Only problem is, you pay 2X the international price because of *heavy* taxes >:(
Title: Re: debugging a small amp
Post by: joecool85 on August 10, 2011, 08:13:08 AM
Not really.
It was just an example of what troubleshooting means, not only in Electronics but anywhere else.
H*ck !!, my Father was an Old style Country Doctor and he used basically the same mindset to "troubleshoot" PEOPLE !!   ;D

Back to cars, you have *any* car in the world available here, thanks to "Globalization".
Only problem is, you pay 2X the international price because of *heavy* taxes >:(

Yeah, we get all the "Americanized" versions of things that suck.  Like the Ford Focus until this year was different than the Euro one, now that we have the Euro one it is considered to be the best compact you can get in the US - go figure.
Title: Re: debugging a small amp
Post by: J M Fahey on August 10, 2011, 12:01:11 PM
Just checking out of curiosity.
I found Ford Focus US prices from U$16,500 to 22,700 , depending on choices.
Argentine prices run from U$20,900 to 32,400.
Not that much difference as I remembered, now only 50% up.
It used to be 100%, which effectively protected us, so we had a big car industry, with almost a million jobs (in a then 30 million Country).
Now "nice cars are cheaper" .... but that million ex-workers now lives out of a State Help plan, in a slum .... and massively votes for today's Government, so they are encouraged to keep things this way.
Welcome to the Future. :'(

EDIT: just checked Brazil, which aggressively protects its industry.
There the (Argentine made) Ford Focus costs the staggering U$43,200  :o ... but they have the largest Car industry in Latin America, world's fifth largest (surpassed France in 2008), so they must be doing things right.
Title: Re: debugging a small amp
Post by: joecool85 on August 10, 2011, 12:50:11 PM
I hear you Juan.  Almost everything sold here is made in China, Taiwan or Japan in that order.  And the US Gov't is going to try and make it worse by getting rid of the tariff on shoes from Vietnam which will effectively push New Balance (still employing 1,000 US workers) out of the US entirely.  There use to be a HUGE shoe industry here in Maine and now it is all but gone.
Title: Re: debugging a small amp
Post by: kin0 on August 12, 2011, 01:19:36 PM

*If* N(ode)-R14/C11/R13 is 9V, on N-R15/R14/C12 you *must* have half that, since both R14 and R15 are same value, 10K.(check it)
You do not  :(
C12 might be inverted/upside down (make sure negative goes to ground, positive to Vb)
IC1 might be bad after accidental inversion.
Measure Vb , first without IC , then without C12 if necessary.


I took out the  IC and the voltage at the node of R14/R15 is still 1.7V. I don't see any point in getting out the C12 because it's not connected and it's in the right direction. However there is a problem with the resistor - their color band shows 10k (Brown-Black-Orange-Gold) but while multimeter I measured -13k to -17K (it was jumping all the time). Should I put a new pair in?
Title: Re: debugging a small amp
Post by: J M Fahey on August 12, 2011, 02:44:18 PM
Quote
I don't see any point in getting out the C12 because it's not connected and it's in the right direction.
Well, *is* it connected or is it not? ;)
Just humor me and pull it, then re-measure.
Also lift *one* end of R14 and R15 and measure them "in the air"
Also follow any copper track that's connected to the Node-R14-R15 , it might be connected (by error) to something else, by a solder drop, a too carbonized residue, whatever.
Current just doesn't dissapear in the air, there must be a path somewhere.
If you have 9V across 2 equal resistors in series, you must have 1/2 that across each of them.
You do not?
Then put your Sherlock Holmes cap on, grab your trusty looking glass and start searching for the truth.
The pipe is not necessary. ;D
Title: Re: debugging a small amp
Post by: kin0 on August 13, 2011, 01:01:38 PM
Pulled the cap out and the voltage on the R14/R15 node is 3.12V  ;D and the voltage on the other part of the R15 (the one that goes to ground) is still 0V. Oh, and I lifted the both R4 and R15 legs and both resistors are exactly 10K. And the traces don't connect to anything that they don't have too
Title: Re: debugging a small amp
Post by: J M Fahey on August 13, 2011, 11:55:19 PM
OK. Node R14-R15 3.15V .................. Fine.
Node R15-Ground 0V ........................ Fine.
And "the other end" of R14 ? :trouble :trouble :trouble
Title: Re: debugging a small amp
Post by: kin0 on August 14, 2011, 01:57:16 AM
still the same 9v...
Title: Re: debugging a small amp
Post by: J M Fahey on August 14, 2011, 02:07:02 AM
OK, lift the R15 leg that goes to ground, and re measure voltage to ground at the R14-R15 node and on the other side of R14. (2 values)
Title: Re: debugging a small amp
Post by: kin0 on August 14, 2011, 07:48:02 AM
oh I just connected R15 back to its place  xP
Title: Re: debugging a small amp
Post by: J M Fahey on August 14, 2011, 09:19:17 AM
And the voltages would be ? .........
Title: Re: debugging a small amp
Post by: kin0 on August 14, 2011, 10:13:54 AM
I was checking the ohms and read your post 10 minuets later. I'll check it then I'm back home
Title: Re: debugging a small amp
Post by: kin0 on August 15, 2011, 05:42:08 AM
Okay

R14/R15 node is 2.25V
the other side of the R14 is 9.12V and the other side of the R15 is 2.01V
Title: Re: debugging a small amp
Post by: J M Fahey on August 15, 2011, 11:59:20 AM
Quote
and the other side of the R15 is 2.01V
8 posts above you said "the other side of R15 is 0 V"
Yes, I asked you to lift a leg to measure it resistance "in the air" but then you 
said "just connected R15 back to its place".
Which one of those contradicting ones is the correct answer?
Besides that, at the R14/R15 node you *must* have 1/2 the voltage you have across both in series.
MUST, period.
*If* you do not, then SOMETHING ELSE IS CONDUCTING CURRENT TO GROUND.
You have the board, you check it, with good light and a loupe, if necessary.
If the board is homemade, you may have improperly etched copper connecting what it shouldn't or small slivers of copper when you perforated it or carbonized flux acting as an unexpected resistor or a blob of solder touching what it should not.
You have the board, you find it.
We can't go on without solving that.
I assume the IC is not there, is it?
Just for you to understand this: solder two new, free 10K resistors "in the air", forget the board, and connect free ends to a 9V battery.
What do you measure across each resistor?
OK: you should measure the same onboard.
If you do not, (so far you don't), find why.
Good luck.
Title: Re: debugging a small amp
Post by: kin0 on August 15, 2011, 03:38:13 PM
In the last post I meant that then the R15 was in air the part which was connected to node of R14/15 was 2.12V and the other part which was in the air was 2V. I'll try to find the problem and post if there will be any improvement.
Title: Re: debugging a small amp
Post by: phatt on August 15, 2011, 06:49:51 PM
Have you checked your DMM leads lately?

Have you checked the battery in the DMM?
They can do strange things to your readings when battery is low.

Phil.
Title: Re: debugging a small amp
Post by: J M Fahey on August 15, 2011, 07:13:12 PM
Fine. As you can see  you have a 10K resistor connected to +9V, its other end connected to another resistor "in the air" so it has a gazillion ohms (open circuit) SO: what's pulling your voltage down?
That's what you have to find.
Do not discard *anything*.
Maybe you say "the track ends in an IC socket , there's no IC in there so the current can't leak there .... "
Wrong.
How do I know?
Well , really I don't, let's check.
What I can't say is "no, everything's right, there's no problem there".
Good luck.
Title: Re: debugging a small amp
Post by: kin0 on August 20, 2011, 01:22:40 PM
I checked all the things that are close to the ground of the pcb and none of them is connected to it accidently. Also I checked all the points and none of the points get electricty (except the ones that need).

What else can I check? (I really have no idea so ANY cluse would be welcomed)
Title: Re: debugging a small amp
Post by: kin0 on August 22, 2011, 10:53:15 AM
Someone please  :(
Title: Re: debugging a small amp
Post by: joecool85 on August 22, 2011, 12:40:06 PM
Sounds like your resistor pair aren't made up of the proper resistors (R14/R15).  Double check their values.  Maybe even replace them with known correct resistors just to get that out of the equation.
Title: Re: debugging a small amp
Post by: phatt on August 23, 2011, 10:21:50 AM
Hi Kin0,
         Sorry to hear you are still struggling with this. :'(
At this point you really need to take close up pictures of both sides of the board as that will help a lot.

Simple truth is that (Even with pics) it is sometimes impossible to trouble shoot circuits over the net like this.
Cheers, Phil.
Title: Re: debugging a small amp
Post by: kin0 on August 23, 2011, 03:14:10 PM
here are 2 pics but the soldering looks much more dirty then in reality-maybe the flash just shows staff that I can't see regularly or just gives bad light.
Title: Re: debugging a small amp
Post by: phatt on August 24, 2011, 09:22:29 AM
Hi Kin0,
Good pics. :tu:
From what I can see You've connected pin 4 (which is circuit com) to the bias rail via a bridging wire. This effectively shorts out the bias voltage to ground.  So hence it won't work.

Try removing that wire.
The Com track is right beside the bias track as it passes under the chip, easy mistake to make.

I realize that etching your own boards is not easy which is why I tend to avoid it where possible.

For my home builds I rework the pcbs and stretch out the tracks a bit.

These kind of pcbs are really only useable IF you have the right equipment to print, etch and drill the boards.

I hand drew a complex pcb once,,,which did work but never again. xP

Hope it works now.
Cheers, Phil.
Title: Re: debugging a small amp
Post by: kin0 on August 24, 2011, 11:33:19 AM
It is my first pcb so i f*cked up some things but I the next one should be better (btw in this one the tracks were to thin for home etching). I'll try to do the thing you said today and post what happens.
Title: Re: debugging a small amp
Post by: kin0 on August 24, 2011, 04:12:41 PM
Oh, I tried to find now the wire you talked about but I have a hard time. Can you circle where it is in the picture?
Title: Re: debugging a small amp
Post by: J M Fahey on August 24, 2011, 06:21:59 PM
Hi Kin.
*IF* ou have the ready to transfer PCB design, the pure and clean bottom side, (you posted a "transparent" view from the component side, if I'm not mistaken), I'll make one showing the step by step iron on process, plus etching, drilling and fluxing; I think it might be useful.
Title: Re: debugging a small amp
Post by: kin0 on August 25, 2011, 02:56:48 AM
of course I have it. But still can phatt point me to the wire he talked about?
Title: Re: debugging a small amp
Post by: phatt on August 25, 2011, 11:04:09 AM
There is only *One bridging wire* you added that passes under the opamp.
It looks as if it shorts the bias to ground. remove it and see if it fires up.

You have 3 tracks under that chip, 9VDC, Bias and Ground.
Phil.
Title: Re: debugging a small amp
Post by: kin0 on August 25, 2011, 03:54:07 PM
Oh, this one. I'll take it out.
Title: Re: debugging a small amp
Post by: joecool85 on August 29, 2011, 08:45:30 AM
Oh, this one. I'll take it out.

How'd that turn out for you?
Title: Re: debugging a small amp
Post by: kin0 on August 29, 2011, 09:11:24 AM
Oh finally the R10/14/15 node is 4.6v!!
Now i need start putting things back in place without lowering the voltage. How this can be done (especially with the cap).
Title: Re: debugging a small amp
Post by: kin0 on August 30, 2011, 04:41:32 AM
I inserted the C12 and connected it to where it needs to be connected and it's still was 4.6V, how ever then trying to connect the R14/15 node to R11/4/5 node it went down to 1.2V. I think the problem is in this node so I'll take out the wires which I used to recreate the track and try to find another way to recreate the tracks.
Title: Re: debugging a small amp
Post by: J M Fahey on August 30, 2011, 08:25:15 PM
DearKin.
A few posts above I offered to make a board at home, even to document the job.
Problem is, when I checked the design you so kindly posted, I felt dismayed.
It's a relatively difficult board to home make  :(
I'm not saying it's a *bad* design, it can be commercially made with no problems. but for home made toner transfer, specially using not the real PNP but some ersatz equivalent, such as magazine paper ... ugh !!
Fact is, traces are unnecessarily thin, much worse rge pads are too small, with relatively large holes; that combination leaves a thin metal ring around component legs.
To further complicate things, the (unnecessary) ground fill areas, complicat the paper removal.
If it's a photo-board or a silkscreened one, no problem, but with thermal transfer ....
In fact I started to redraw it with twice thicker tracks and pads, using my trusty Protel Autotrax 1.61 freeware, but really don't have the time so I quit.
I also found a design error: Pin1 of IC1A sits at 4.5V DC , through R5 and R9 it applies DC voltage to the tone pot, you end having over 1V DC across it.
Instant scratchy pot, big way.
You should replace R6 with a 1uFx16V electrolytic, positive pointing left.
The range of the "Tone" pot is very subtle.
Maybe it's really that way, or it was improperly copied.
Oh well.
Title: Re: debugging a small amp
Post by: phatt on September 01, 2011, 09:57:14 AM
Thanks Mr Fahey,, You have given me some ideas to work with as I do find Eagle a pain. :tu:

I did have a copy of  Autotrax but was running Win Me at the time and it refused to work. :'(

Phil.
Title: Re: debugging a small amp
Post by: J M Fahey on September 01, 2011, 12:31:21 PM
Hi Phil of Nambour Australia.
Please let me remind you that using any software besides a Protel one (Easytrax or Autotrax are acceptable) will instantly strip you of your Citizenship.
You will be lashed, not with a 9 cat tail whip buy with an actual cat.

I hate Eagle, simple as that.

*Up to* XP you can run Autotrax if you start the machine in DOS mode *or* in software properties state parameters as DOS would. No "automatic" memory settings (it does not know how to ask for it) but you reserve the maximum for it in each little window *or* download the excellent DOS BOX program which takes a little setting up but then does everything for you *or* use any old machine you have rusting in the garaje .
 Win 3.1/95/98 will do, in any old hard drive (even a 500 Mb one).

I am running it in Dos Box in a Windows 7 netbook and "printing" to a PCX file, which later any graphics software can handle.

Something I forgot to mention, a modern program which does not need so many tricks to run:
I will experiment with Express PCB which is very similar in concept, "paper and pencil drawing board", and is available for any Windows.

It has no size or pin limitations, but being a proprietary program forces you to order the boards from them.
As you already know, there is not such a thing as a free lunch.
Anyway, for homemakers you can print the PCB design (using a little trick) and thermal transfer it.
Title: Re: debugging a small amp
Post by: phatt on September 11, 2011, 10:09:41 AM
Thanks Juan,

Yes I do now have an old clunky win98 machine and I might just have a fiddle with it and see what happens.

I did get a patchy grip on DOS quite some time back So I do have some ability to get things to work with older gear.
Lets just hope the old machine still works. (Yes it is in the shed collecting dust)

My main excuse for not getting boards to look pretty is I hate all the mucking around with printing the boards ready for etching.

I'm just using a pen and hand draw as I'm not in production I don't see the point in outlaying money to make sexy boards that no one will ever see. As long as they work it's valid for me.

Most of the stuff I've made is fairly simple by hand but now that I'm trying to bring a few ideas all into one unit I may have to rethink my whole approach.

Eagle just helps to get pins for opamps and like stuff in the right place.
Then lightly centre pop the holes, the rest is just joining the dots with an ink pen> etch> drill> done.

Of course,,,More than one or two opamps,,,you go nuts trying to keep track of all the dots.

Meantime time I'll keep an eye out for a strange man walking up my street with a cage full of angry cats.  :P
Phil.
Title: Re: debugging a small amp
Post by: J M Fahey on November 06, 2011, 04:44:07 PM
The real problem starts when he throws on you a bucket full of old fish ... and then leaves the cats free. :lmao:

You can do the exact same thing with Easy/Auto-Trax.
I used to test designs by printing the PCB artwork on regular paper, any printer will do, even a prehistoric Ribbon Dot Matrix one, tape said artwork over the copper board with a sheet of carbon paper (incredibly still available) in the middle, trace the tracks with a ball point pen (which leaves a trace on the copper) , pin through all holes with a sharp needle or similar, and then it´s very easy to handtrace with Sharpie pens.
The faint trace on copper helps a lot .

Now it´s faster and easier for me to thermal transfer, practice makes perfect.