Em Drive clone not working properly

[Grubb]

Hi folks,

Having searched the forum, I believe I may have made a common beginner's mistake: setting out to make an Em-Drive because of their low parts count. I now realise that they're not a great design. But I intend to at least try and get what I have made working, having put a fair amount of effort into it so far.

I took a schematic I found online and added some basic things to it (e.g. power circuit stuff like a polarity protection diode) and this is the first place I may have made an error - if we don't include choosing an Em-Drive to clone :


Then I worked out a layout in DipTrace, which may also be problematic:




And then I had the board fabricated at OSH Park, just to see if I could actually make a working PCB. Now I have populated the board and wired it up to a PedalPCB Auditorium platform for testing. The LED lights up, there is sound, but it sounds pretty awful, like a spitty, gated fuzz with terrible sustain and inconsistent control from both pots. I figured the issue was the internal trim pot not biasing the MPSA18 properly, so I played with it manually, but to no avail. Then I did some research and realised I probably needed to post some voltages and get input from people who know what they are doing.

Q1
C: 0.123V  <--- I'm guessing that's too low?
B: 0.638V
E: 0V

Voltage at 9V+ board input: 9.2V
Voltage at 9V- board input: 0

The transistor is oriented correctly, I have checked and re-checked. Suggestions? Thanks for your expertise 

[antonis]

Hi & Welcome..

Yes, Collector voltage indeed is TOOOOOO low..
(propably due to false bias configuration)

Try to move R1 leg 2 to TRIM1 leg 1..
(Collector - Base NFB bias..)

[GGBB]

The schematic is probably wrong - pins 1 and 2 of the trimpot need to be connected to each other - this is also missing from the pcb. This is preventing the bias control from working.

Also - on the pcb - there are no connections to ground or "+" - are you using top and bottom planes for those? I assume so since the LED works, but please confirm.

EDIT: You should be able to fix the trimpot issue easily by soldering a jumper between the two pins on the back of the pcb. You shouldn't need the fix suggested by Antonis, but it wouldn't hurt.

[Grubb]

Thanks for the advice 😊

I'm not all that surprised that the schematic and PCB turn out to have mistakes in them. I will try the jumper and if needs be I could rig up the resistor differently too.

Yes I used planes for ground and 9V+ although I would do that differently now if I was doing it again. I've also started changing trace sizes to be larger for power and smaller for signal traces since I did this board.

Do the marks on this schematic (done clumsily on my phone) represent the changes you are both suggesting?

[GGBB]

Do the marks on this schematic (done clumsily on my phone) represent the changes you are both suggesting?

Looks good.

[Grubb]

I just ran the jumper and adjusted the bias to around 5V and it sounds pretty good as a subtle boost/drive. I wouldn't want it adding any more than a touch of hair, the clipping at higher gain isn't a nice sound. Thanks for helping me get it working! Now to finalise the bias and box it up 😊

[anotherjim]

If the power arrangement in the schematic is what you built then...
Does the battery ever switch off?
When you plug 9v power in, is it still left to the battery to power the LED?
If the trim pot is turned fully in one direction it can be 0ohm and that will probably destroy Q1 or damage the trim pot - or both.

[Grubb]

Hi Jim, I don't use batteries in any of my pedals, not even in the ones I buy. It's just been tested with a 1Spot 9v power supply.

I did have the trimpot at both extremes at various stages and thankfully nothing seems to have broken permanently so far. It's been a good learning experience, even if I'm painfully aware that I don't really know what I'm doing yet.