Need help on a Idiotbox-Blowerbox

[FredG]

Hello Diyers

I build an Idiotbox-blowerbox but I get no sound from it.
When it's off, bypass is ok.
When it's on, led is ok, but no sound at all.
(and I did not mistake the in an out jack as I do all the time, it's always my first debugging step  )

- Problem (?) : I have a connection between these 2 parts of the same line (in red). At first I was thinking of a solder bridge, but there is not.
And when I disconnect the 1k resistor (enhanced on the schematic picture), this connection disappears.
Maybe it's supposed to be that way
I'm a beginner, but I red many times that there should never have a connection between two separated part of the same line.
But I don't understand schematics so I need your help.

Or maybe I made a mistake somewhere, despite the fact that I triple checked every component's position, cuts, wiring, etc ?

Tank you very much in advance for all the help you can give.

Here is the layout page : http://tagboardeffects.blogspot.com/2014/07/idiotbox-blowerbox.html
Here are all the pictures :

[Kipper4]

You need to help us to help you here mate.
Can you find a schematic for that.
Please share you're voltages too. This may help.

Also read this

https://www.diystompboxes.com/smfforum/index.php?topic=29816.0

[Kipper4]

This might be it.


http://johnkvintageguitars.homestead.com/Effects/Fuzz-ODs/Idiotbox/BlowerBox_Schematic.png

[FredG]

Yes, this also the schematic I found.

And here is the voltage (thank's for the link to the debugging page by the way... I still have a lot to discover in this forum) :

Board V+ = 9,63
Board V- = 0

Q1
D 7,51
S 0,25
G 0,01

IC1
P1 8,07
P2 0,09
P3 0,70
P4 0
P5 0
P6 7,62
P7 9,02
P8 8,76

D1
A 0
K 0,17

D2
A 0,1
K 0

D3
A 9,62
K 9,63

The only piece that is different from the layout is the treble pot that is not reverse (still have to buy one).

Thank's again. 

[amptramp]

1. The LM308 is an interesting choice here - it is designed as a high DC accuracy op amp that is designed for a load of 10K but the DC accuracy is not used here.  It also features high input impedance, so it was better than many FET input op amps at the time it was designed.  If you have the gain all the way down, the midband load is set by the 47 ohm resistor going to the 2.2µF cap and the bass load is set by the 560 ohm resistor going to the 4.7µF cap.  If you turn the 50 K gain pot below 10 K, the op amp will not like the load and may just quit.  Check if it behaves at high gain.  BTW this device is slightly above 30 nV / root Hz under even the best conditions, so it will be a little more noisy than necessary.

2. There needs to be a bypass capacitor across the pin 7 / pin 4 power rails of the op amp.

3. The 1 nF cap from the op amp pin 3 to ground is about ten times the value usually used.  This is to filter out incoming radio signals so they do not upset the bias point, even if you don't hear these signals directly.

4. Check if you have an output at the diodes.  The output of the op amp can only be tested with a high impedance probe if you don't want to upset it but the isolation provided by the 1 K resistor should be enough to keep it from oscillating.  Taking a line from this point to the guitar amp input may show signal there and if it does, you problem is downstream, if not, it is in the LM308 stage.

5. What is most likely to be the problem is if the schematic is correct, there is no gate return for the JFET.  The gate voltage is free to float up or down and if down, it will cut off, if up, it will pull the JFET into full conduction.  Maybe the 1 meg resistor across the output pot was intended to be connected to the gate of the JFET instead.  As it stands, there is no gate return, so this voltage will be set by leakage currents and will probably turn the JFET full off or full on.

*****

Check item 5 first!  If you put 1 meg to ground from the gate and it works, leave it there.  Another possibility is to short out the 10 µF going to the gate of the JFET.  This biases the JFET through the tone controls, but biasing the device through a pot may be noisy.  Good luck and good hunting!

[reddesert]

The Blowerbox is essentially a Rat with a Baxandall / James bass/treble tone stack. That is why it uses an LM308, although other single op-amps like a 741 will work.

Your DC op-amp voltages are weird. Like most pedal circuits, the Blowerbox sets the DC bias of the op-amp to about half way between Vsupply and ground, so about +4.5V.  You can see on the schematic how this is done: the two 100K resistors in the upper left are a voltage divider between Vsupply and ground, so their junction is at about +4.5V. And the only DC connection of the op-amp inputs is from pin 3 through a 1M resistor to this 4.5V point. All of the other paths from an op-amp input to anything are DC-blocked by capacitors.

Thus, when measuring DC, pins 2 and 3 should be at about +4.5V. Pin 6, the output, will also be at 4.5V because its only DC reference is through the gain pot to pin 2.

There is likely a short, an open connection (like an unsoldered joint), or a wrong-value resistor, somewhere around the op-amp or in the voltage divider that is supposed to make +4.5V.  Until you get that fixed, the op-amp will not amplify, as its output is pinned to one of the supply voltage rails.

[reddesert]

5. What is most likely to be the problem is if the schematic is correct, there is no gate return for the JFET.  The gate voltage is free to float up or down and if down, it will cut off, if up, it will pull the JFET into full conduction.  Maybe the 1 meg resistor across the output pot was intended to be connected to the gate of the JFET instead.  As it stands, there is no gate return, so this voltage will be set by leakage currents and will probably turn the JFET full off or full on.

*****

Check item 5 first!  If you put 1 meg to ground from the gate and it works, leave it there.  Another possibility is to short out the 10 µF going to the gate of the JFET.  This biases the JFET through the tone controls, but biasing the device through a pot may be noisy.  Good luck and good hunting!

Excellent point. In the vero layout shown from tagboardeffects shown above, the 1M resistor goes from the gate of the JFET to a strip that is ground (it's linked to pin 1 of the Volume pot).  The vero layout is right and the schematic is incorrect.

I've built a Blowerbox on the tagboardeffects vero (as have many others), it does work.

[FredG]

Thank you very much for these answers and all the explanations.   
I will first check all the values and possible shortage around the IC and voltage divider.

=> And, going back to my first question : is it supposed to have a connection between the 2 parts of the same strip that I enhanced in red in the first picture ? (Connection that disapears when I disconnect the 1k resistor)

[patrick398]

=> And, going back to my first question : is it supposed to have a connection between the 2 parts of the same strip that I enhanced in red in the first picture ? (Connection that disapears when I disconnect the 1k resistor)

When you say connection do you mean continuity on your multimeter? I just can't see how those two parts would be connected except through the 10uf and 100pf capacitors but that would require them both to be bad.

Did you check for no continuity between tracks after you made the cuts? (before you populated the board)
They look pretty decent from the photo you posted so maybe not the issue.

Actually, are you testing for continuity with the gain pot all the way up or down? If that 10uf cap is bad i think the pot being all the way up or down would create continuity between those two points.

[FredG]

Yes, I meant continuity (sorry, I have to learn english and electronic at the same time, my brain is not really designed for this  )

I tested the board before populating it and it was ok.
I also cleaned between each strip with a blade after populating it.

The continuity is there, regardless of the gain pot at min or max or in between.
Ther is also a continuity between the center part of the same strip and the left part.
And sometime the continuity happens for 1 or 2 seconds, then disapears, and sometime it stays... 

[antonis]

And sometime the continuity happens for 1 or 2 seconds, then disapears, and sometime it stays... 

You may call it "cold joint", if you like...

[FredG]

Ok, now the problematic short is settled (the second 47ohm resistor on top left was not in the right spot).

But still no sound coming out when on.

Here are the new values I get :

Board V+ = 9,5
Board V- = 0

Q1
D 6,9
S 0,2
G 0

IC1
P1 8,4
P2 7,8
P3 7,2
P4 0
P5 0
P6 7,7
P7 7,8
P8 8,76

D1
A 0
K 0,1

D2
A 0,1
K 0

D3
A 9,5
K 9,2

I checked every resistor values, they are all correct.

@Redesert told me that the IC's pin 2, 3 and 6 should be around 4,5 . but they are now  around 7,5...
As I have now no more short, maybe these significant differences between what I have and what it should be can help you to diagnose the issue ?

Thank you very much in advance for any comment and help.

[PRR]

Missing ground around the power protection diode and the two 100K divider resistors.

[FredG]

Thank you PRR.
But I am a beginner and from what I see they are all connected directly to the ground strip on the left. Does it mean that the 100uF cap is not doing the job, or...?
What do I have to look for ?
Thank's again.

[PRR]

I suspect bad joints. Looks like it connects, but doesn't really.

Look close at your (excellent) board photo.


There is lots of solder on the board copper. But I see "black rings" around resistor leads. This may be burnt flux, which is not a conductor. The solder didn't actually "wet" the resistor lead. Prime cause is not enough heat on the lead. Also the resistor leads are usually tarnished in storage, don't want to take flux and solder.

Many other joints are so blobbed-up with copper that I can not see the lead, can not even guess if there is any solder actually touching and "wetting" lead metal.

This is the bane of electronics. Solder can be a good connection. But EVERY joint has to be GOOD. One of the main incentives for developing chips in the 1960s is that a million-solder-joint computer was a headache to get working; more parts per chip meant less solder joints.

IMHO: heat those joints (any that you can't eyeball as GOOD), beat-off the solder. Now scrap the lead end, use a hot iron and good flux solder, tin just the lead-end and eyeball it as totally fresh tin. Then as a second step, heat both lead and PCB, touch solder, get a thin WET bond to both PCB and lead.

[PRR]

This joint looks familiar. We had a PC monitor which would "go out". A smack usually got it going for a while. But it got worse. When my hand got tired I opened it up. Where the main HV lead was soldered to the PCB, there was a flux ring 3/4 of the way around. (Red in my mark-up.) Originally there was a fine connection through the 1/4 which really wet the lead. (Orange in my mark-up.)


But it was a long lead and vibration broke the thin sliver of solder. Even then it often worked because lead tension tended to hold the pieces in-contact. But rust never sleeps, and people bump desks, and the connection was not reliable. Any final inspector would reject a joint which had more than a tiny pinhole of no-solder. If the joint is otherwise good and life will be easy, sometimes a half-diameter solder joint will work forever; but I only do that for myself, not something to go Out In The World.

[FredG]

Wow !
Thank you very much for these explanations. Very very usefull ! 

[FredG]

Yes ! It finally started ! (Bad soldering was the final issue with this boxe)

Now I can proudly make noise with this feeling that it sounds even better when I had to sweat on it for several days. 

Thank you very much to every one who helped me here.
I learned a lot on this project, thanks to you guys !