Please help me out - debugging my Ruby Bassman amp

[Rutger]

Hi,

I posted my topic in another forum but got the advice to post it here, for there are more people to help me out...

http://www.diystompboxes.com/smfforum/index.php?topic=84304.0

As you can read, my Ruby amp doesn't work and I measure far to less voltage on P5. Can anyone tell me how to fix it?

Thanks!

[Brymus]

Like Gil said in your original thread your FET isnt biased right for some reason.

[Rutger]

I figured out that the JFET was broke and installed a new one today. Checked the values of the resistors and the solderjoints again before I fired it up.... but I still got 0.7V on P5 of the LM396.... does anyone have an idea what it could be?

[PRR]

> IC
> P1= 2,9V (gain on 10)
> P2= 0,4V
> P3= 0V
> P4= 0V
> P5= 0,7V
> P6= 9,2V
> P7= 0,4V
> P8= 4,6V (gain on 10)

It looks like a short from pin 7 to pin 2, plus a shorted cap from the JFET to pin 2.

Expected:
P1= 1.2V
P2= 0.0V
P3= 0.0V
P4= 0V
P5= 4.6V
P6= 9.2V
P7= 5.2V
P8= 1.2V

By the way: this is LM386, -not- "LM396". I'm sure you did not use the '96, because you can hardly find one, and it only has 2 legs.

[Rutger]

Sorry, that was just a typo 

Thanks alot for your help, I'll take a look at it tonight!

[Rutger]

I have my LM386 problem fixed. Quite embarrasing actually... I soldered the layout looking at the back of the perfboard, so that the layout is mirrored when you look at the parts on top. That's no problem for most of the parts, but you can't mirror the IC ofcourse... so I reversed the legs of the IC and placed it back into it's socket, now Its working like it should! I measure the right voltages and it actually works when I skip the JFET.

Only thing now is the JFET. I checked everything over and over, solderjoints, parts, layout, etc... I still measure 9,2V 8,4V and 8,4V at D, S and G with a new JFET. Maybe the configuration of D,S and G differs from standard configuration...? I'll test the JFET's again tonight.

I keep staring at the darn thing, sometimes it's so obvious that you keep overlooking it.

[PRR]

[Rutger]

Yes I know

Well, this mirrored soldering thing kind of freaked me out... turns out that I made the same mistake with the JFET as with the LM386 over and over... I had it reversed.    
It seemed right when I looked on my layout and at the back of the perfboard, but I should 've looked at the top of the perfboard to see I was doing it the wrong way.
The pots same thing: they are all reversed so they work anti-clockwise instead of clockwise  

Personal learning point: never, NEVER solder you layout from the back of the perfboard! It messes with your head.

Well, the positive thing about all this is that I learned how to debug a circuit and check everything in detail befor I do anything stupid. It's my first project ever, and it's nice to know that there's nothing wrong with my electronics or soldering skills. Only with my common sense, haha! It gives me confidence to go further to a next project

Now it works, and works very good. The noise level is very low. It does what they tell about it. Cleans are really nice, it's quite responsive, the overdrive is okay. Just a typical stompbox kind-a overdrive, nothing tube-like really. I prefer the 2N5457 over the MPF201, just a subtle difference. I placed a 33k resistor right before the JFET for a tighter overdrive and it seems to do it's job nicely. The tonepot actually works quite nice as well, far more subtle than most toneknobs in stompboxes.

Thank you all for your tips and patience!

[PRR]

> there's nothing wrong with my electronics or soldering skills. Only with my common sense

It is often the "easy" "obvious" things which make trouble.

> the positive thing about all this is that I learned how to debug a circuit and check everything in detail

That's a very positive thing.

[stringsthings]

Yes I know

Well, this mirrored soldering thing kind of freaked me out... turns out that I made the same mistake with the JFET as with the LM386 over and over... I had it reversed.    
It seemed right when I looked on my layout and at the back of the perfboard, but I should 've looked at the top of the perfboard to see I was doing it the wrong way.
The pots same thing: they are all reversed so they work anti-clockwise instead of clockwise  

Personal learning point: never, NEVER solder you layout from the back of the perfboard! It messes with your head.

Well, the positive thing about all this is that I learned how to debug a circuit and check everything in detail befor I do anything stupid. It's my first project ever, and it's nice to know that there's nothing wrong with my electronics or soldering skills. Only with my common sense, haha! It gives me confidence to go further to a next project

Now it works, and works very good. The noise level is very low. It does what they tell about it. Cleans are really nice, it's quite responsive, the overdrive is okay. Just a typical stompbox kind-a overdrive, nothing tube-like really. I prefer the 2N5457 over the MPF201, just a subtle difference. I placed a 33k resistor right before the JFET for a tighter overdrive and it seems to do it's job nicely. The tonepot actually works quite nice as well, far more subtle than most toneknobs in stompboxes.

Thank you all for your tips and patience!

congrats on debugging the problem ! ..... we've all been there .... i recently got back into DIY after a long layoff and i smoked a chip by hooking up the power supply backwards !  ..... you'll find that debugging is just another part of the hobby ....

tip:  when a circuit isn't functioning properly, take notes on the problem .... record sound samples if possible .... and take frequent breaks to avoid frustration

[Rutger]

A week further now, and still happy with this little amp. I have one problem left, and that's the mastervolume control. I placed a 100k-A pot right before the headphone/speaker-out, like in te runnoffgroove Big Daddy, but it's not working properly: In the first 40% of the turn there is no signal, then it acts like a volumepot untill about 90%, and then suddenly it switches to maximum output.

Well, since I placed the mastervolume especially for use with headphones, the range where the mastervolume works properly is good enough since the volume is loud enough trough headphones. But it isn't very pretty... So I was thinking, is it a good solution to replace the 100k-A for a 50k-A, and then place a 50k resistor from pot to ground?

Besides that, can anyone explain this weird behaviour of the volumepot to me? Does it have something to do with the circuit, or how a voltagedevider behaves?
Makes you think that it'll not be working with other designs like the Big Daddy either...

[PRR]

> 100k-A pot right before the headphone/speaker-out

You can't drive a 8 ohm speaker or 32 ohm phones with a 100,000 ohm pot.

Picture the 100K pot at 50% electrical setting. 50K from wiper to ground. 50K from wiper to top. This is a voltage divider. The UN-loaded output is 50%.

Now hang 32 ohms from wiper to ground. This is in parallel with the lower 50K. The result is 31.98 ohms.

The top of the pot is still 50K.

The voltage divider ratio is 31.98 / (50,000+31.98). The output is 0.000,64 times the input. 63dB down. Not literally "no" output, but too weak to hear well, if at all.

The pot value should be "similar" to load value. For "exact accuracy", much lower than load. In audio, we turn-to-taste, so we don't have to use a lower value.

In audio, it is often effective to use a somewhat high-value linear pot. The loading at mid-rotation gives a quasi-"audio" taper in the middle of the dial.

Try 100 ohms. (May be hard to find.)

Try a "loudspeaker attenuator". In the US we can find wall-mount volume controls at Radio Shack. They are intended to control ceiling speakers in offices and meeting-rooms.
http://www.radioshack.com/product/index.jsp?productId=2104029
The $20 price is a bit silly. Especially since you will throw-away the wall-plate. Careful shopping at low-price PA supply stores should find something similar, cheaper.

> like in te runnoffgroove Big Daddy

That page may not say, but it is intended to go into a Guitar Amp. Such inputs are usually 100K-1Meg. The 100K pot is suitable for such high impedance loads.

Also the 0.1uFd cap shown on that plan is ample for 100K+ loads, would be incredibly weak and gut-less for 8 and 32 ohm loads.

[Rutger]

I see, thank you very much for this clear explanation!

Yeah, those pots are very hard to find...maybe I can just place a small resistor to ground and make it switchable.

[Rutger]

Hi,

Since I made the Ruby work it had been laying around in a box since I haven't had the time to build it in a housing untill now. Everything is installed now and since my mastervolume didn't work, I replaced it with a switch to ground so I can switch in a little more overdrive.

But now it farts, that arteficial overdrive kind of noise on top of the guitarsound that appears when you pick the strings a little harder, then stays on when the notes die out and finally suddenly dissapear. It farts in every setting, clean and overdriven, especially the low notes.  First I thought I had to look at the last change I made (the switch) but bridging the switch doesn't solve the problem. I have no clue how to solve this, so can anyone tell me where I should start looking...?

[Rutger]

Anyone?

[Brymus]

It should clean up if you disconnect pins 1 and 8 ,this will set the gain back to 20
If that doesnt work check how hot your input FET is biased.
Or just use an adio probe or just take the signal right after your FET and connect to your amp,
The signal needs to be fairly clean going into the 386 to start with,and the 386 is easily overdriven at the input.
Which gives it that nice break up.

[Rutger]

Thanks

I found a loose solderjoint and fixed it, from input to earth. Didn't fix the problem but gave me a very sensitive input so it picks up every bit of noise and a German radiostation...  
Seems like every time I get my hands on this little build some new problem appears  

Measured the voltages of the JFET and IC again. The JFET is fine, but I found a slight difference on pin 7 of the LM386: 4,6V instead of the 5,2V I've measured before. I must have overlooked it last week. The rest is still fine.

[Top Top]

The .047 cap and 10Ω resistor from the output to ground (going off memory here so sorry if I have those values off) help cut out farting/weird distortions and oscillation, so you may want to check to make sure those are wired correctly and with correct values.

[Rutger]

Thanks, I'll check that today and also the 100n cap from P7 to ground. I haven't be too carefull with it so maybe there's a loose connection there as well.

[Rutger]

This weekend I had some time to take a look at my little project again.

I swapped out the 'gain'switch for it wasn't very useful. I hoped that would do something good about the weird buzz but it didn't. There is obviously something wrong after pin 5 and/or 7, for the cap and resistor to ground are there to prevent such noises. But didn't figure out what yet... any idea how to measure the quality of a cap with a simple multi-meter?

I placed a shielded wire between input and the perfboard to eliminate the noises and radiostations. But again that didn't help much. So I decided to let the electronics for what it is and finished the speakercab for the 8 incher I was working on as well. I installed the speaker in the cab and did a little test run with the ruby connected... noises and radio broadcasting belongs to the past...! No idea how but apparently the speaker had something to do with it.

Now that the speaker is installed the little buzz isn't that noticable anymore. Actually the little cab is quite convincing, I made it closed back and 'oversized' and that is noticeable, it isn't boxy like many practice amps are. It sounds very warm and the speaker still needs to break in. With the ruby it is a joy to play with! The overdrive of the ruby a lot more confincing than through headphones.

I recommend installing the toneknob of the noisy cricket, with the ruby this simple circuit works very subtle and effective. Much better than on an avarage stompbox.

Maybe a little weird to be that enthusiastic about a project like this. But the fact that the project really does what you had in mind before you started building is very satisfactory.
I will use this one as a practiceamp for my students.