Help with pedal switch please

[sonolink]

Hi,
A friend brought me a JHS Bonsai guitar pedal. The pedal works fine but the LED stopped working. It's a pedal emulating several versions of the Tubescreamer.
So I opened the pedal with the idea of replacing the LED. After desoldering it from the board I checked it and surprise! it works perfectly.
So I took a peak at the board and noticed that instead of the config:
9vdc -> current limiting resistor -> LED -> switch -> GND
There's a transistor. My guess is that the transistor is acting as a switch and for some reason is toasted, that's why the LED is not working. But I don't understand why there's a transistor there and if it has any function other than controllong the LED.
I've tried finding a schem, I emailed JHS technical support and even messaged Josh Scott (the CEO) through Facebook. I've had no answer whatsoever and couldn't find a schem.

Could someone help me out please? I've attached a pic of the part of the board where the LED circuit is.
Thanks in advance!!

 

[NoisyIndividual]

I would be looking at the switch first. If it's one of the blue square ones they often fail.

 

[soulsonic]

Replace the switch

 

[sonolink]

Checked the switch. The switch works fine....

 

[ccaudle]

I don't understand why there's a transistor there

What are the connections to each pin of the transistor? Can you tell whether it is a bipolar or MOSFET?

 

[sonolink]

What are the connections to each pin of the transistor? Can you tell whether it is a bipolar or MOSFET?

I hope this helps:

 

[soulsonic]

Are you reading 9,4v across the LED terminals with the LED in place?

 

[sonolink]

Are you reading 9,4v across the LED terminals with the LED in place?

That is correct.

EDIT: No, I read 1.76v across the LED terminals with the LED in place.

 

[soulsonic]

Try shorting the negative lead of the LED directly to ground - which would bypass Q3. If it turns on, you'll know Q3 is dead

 

[sonolink]

Try shorting the negative lead of the LED directly to ground - which would bypass Q3. If it turns on, you'll know Q3 is dead

Q3 is dead....long live the LED!!

Can someone please explain the function of D21 and Q3? Is it only related to the LED?
If so, can I bypass this part of the circuit, feed 9v to the LED through a resistor and use one of the poles of the 3PDT swiswitch for the LED, somehow modding the board?

 

[ccaudle]

My guess is that the transistor is acting as a switch

Yes, it seems to be an open-collector style connection acting as the switch for the LED.

I don't understand why there's a transistor there

The most likely reason to do that is you want a signal to switch multiple things in the circuit, and it doesn't work out to have the LED actually part of those other things, so you use a control signal to turn electronic switches on and off at the same time.

I read 1.76v across the LED terminals with the LED in place

That seems right-ish for the forward voltage drop of a green LED, but on the low side. I just spot checked a couple of datasheets and 2V seems like the expected forward voltage of most green LEDs.
How did you determine that the LED was working? Can you check the forward voltage when it is on?
No other components were previously changed? The main battery voltage stays around 9V when the pedal is switched on, it isn't dropping too low?

Check the voltage across the transistor collector to emitter (LED connection to gnd connection), it should be very low voltage when turned on.

Looks like you figured it out while I was typing.

use one of the poles of the 3PDT swiswitch for the LED

It seems unlikely that JHS would choose a more expensive 3-pole switch if it was not needed. Are you sure you understand what all three poles of the switch are controlling? It's not that hard to change out a 3-pin transistor, why not just replace the failed component?

 

[sonolink]

Yes, it seems to be an open-collector style connection acting as the switch for the LED.



The most likely reason to do that is you want a signal to switch multiple things in the circuit, and it doesn't work out to have the LED actually part of those other things, so you use a control signal to turn electronic switches on and off at the same time.



That seems right-ish for the forward voltage drop of a green LED, but on the low side. I just spot checked a couple of datasheets and 2V seems like the expected forward voltage of most green LEDs.
How did you determine that the LED was working? Can you check the forward voltage when it is on?
No other components were previously changed? The main battery voltage stays around 9V when the pedal is switched on, it isn't dropping too low?

Check the voltage across the transistor collector to emitter (LED connection to gnd connection), it should be very low voltage when turned on.

Looks like you figured it out while I was typing.



It seems unlikely that JHS would choose a more expensive 3-pole switch if it was not needed. Are you sure you understand what all three poles of the switch are controlling? It's not that hard to change out a 3-pin transistor, why not just replace the failed component?

You're probably right but the problem is I don't know what transistor I need...

 

[soulsonic]

You're probably right but the problem is I don't know what transistor I need...

You can just remove that transistor and connect the LED directly to ground. The footswitch should still turn it on/off judging from your drawing of how its wired. The transistor is likely behaving as sort of "soft start" so the LED doesn't cause a popping sound when turning on. I would try it without the transistor just to see how it behaves. And if the LED seems too bright (because of not having the transistor's resistance in series with it anymore) you can increase the size of R62 to compensate. But it appears to be 3K3, which should be enough to keep most LEDs from being too bright.

 

[ccaudle]

I don't know what transistor I need...

MMBT3904 or equivalent would be the usual device for that kind of use, been around for decades, 10 cents in single quantities. The problem is if you don't have one or find someone to give you one, you pay $0.10 for the part and $4.95 in shipping and handling if you have to order one.

There is also a chance it is a 2N7002 MOSFET.

Now that you know specifically what the problem is, try asking JHS what they recommend to replace a damaged Q3. That is a lot more specific question than "pedal broke, what could go wrong?" so maybe more likely to get a response.

 

[NoisyIndividual]

Q3 does appear to have a marking, though it can't be made out from the photos.

 

[sonolink]

You can just remove that transistor and connect the LED directly to ground. The footswitch should still turn it on/off judging from your drawing of how its wired. The transistor is likely behaving as sort of "soft start" so the LED doesn't cause a popping sound when turning on. I would try it without the transistor just to see how it behaves. And if the LED seems too bright (because of not having the transistor's resistance in series with it anymore) you can increase the size of R62 to compensate. But it appears to be 3K3, which should be enough to keep most LEDs from being too bright.

Connecting the LED directly to GND doesn't seem to work. The LED stays ON all the time.

MMBT3904 or equivalent would be the usual device for that kind of use, been around for decades, 10 cents in single quantities. The problem is if you don't have one or find someone to give you one, you pay $0.10 for the part and $4.95 in shipping and handling if you have to order one.

I've got a few 2N3904 around so I strapped on one to check:





The LED goes on but fades out after a second...

Q3 does appear to have a marking, though it can't be made out from the photos.

I tried my best

Now that you know specifically what the problem is, try asking JHS what they recommend to replace a damaged Q3. That is a lot more specific question than "pedal broke, what could go wrong?" so maybe more likely to get a response.

I'll try them again

Thanks a lot everyone for all your ideas and suggestions!
Cheers
Sono