SOLVED CrackleJack - attenuates instead of boosts signal

[jpants]

I really wanted my first post here to be a build report but alas, I require advice. I have some experience soldering on larger surfaces like guitar wiring but this is my first pcb attempt so I'm sure I'm making a lot of beginner mistakes.

I tried to read up and test as much as possible before posting but I'm running out of ideas of what else to test and I'm not sure that I'm making the right conclusions from the tests. So a huge thanks in advance to anyone taking time to read this and any pointers would be greatly appreciated.

I attempted to closely follow the steps outlined here although I didn't really test the circuit other than power to the LED before connecting the switch and jacks:
https://forum.pedalpcb.com/threads/basic-workflow-tips-for-building-a-pedalpcb.1165/

WHAT WORKS:

power, LED and switching work.
bypass signal comes through fine,

WHAT DOESN'T:

bypass signal is louder than effect signal but both are audible.

WHAT I'VE TRIED:

Everything that I tested here I've tested multiple times with a couple days off and some solder reflows in between. The first test was very similar to what I report below which are the results of the most recent tests. Originally, when I found signal with the audio probe everywhere I thought it should be, I assumed that my solder joints weren't good enough and reflowed what I thought were the worst ones. These pictures are all after the reflow, still not very good but definitely better than the first run.

I tested for continuity, I believe it blipped everywhere that its supposed to but I'm not 100% sure I was checking the right parts of the circuit. I double checked the values of all resistors, pots and caps as well as the orientation of all polarized components.

Audio Probe results:
I have signal (almost) everywhere that I think it should be:
- both sides of input cap C1
- top of R1, top of R3
- both sides of R2 and C2
- all 3 pins of the BS170
- both non-ground pins of A100k Volume pot
- but only the leftmost pin of C5k gain pot, should I also have signal on the middle pin?

Is there anywhere else in the circuit that I should be hearing a signal?

Voltage Test: I'm not sure what the correct values should be, but intuitively it seems wrong that I'm losing so much over D1, R1 and R5 since those just go to ground.

Whole circuit: 9.4v measured from power(+) to power(-) pads on pcb
D2: 0.12v
D1: 4.15v
R1: 4.15v
R3: 3.7v
R2: 1.3v
R5: 6.9v

I also don't know what this should be and if I'm interpreting the pinout correctly but here's the voltage across Q1
BS170 (top down with LED up perspective)
Left-Middle (source-gate?): 0.26v
Left-Right (source-drain?): 1.57v
Middle-Right (gate-drain?): 1.3v

What should these readings be and what should I do to fix them? Is the large voltage drop over R5, D1 and R1 indicative of something grounding out that shouldn't be? I have other bs170s and backups of all components if swapping something might help.

I strongly suspect the culprit is my soldering, but I cant figure out which specific parts are the problem or if I've done damage to the board or other components at this point.

Additional newb questions:

Whats up with flux? I couldn't find a flux pen as described in the guide so I used a Jel Flux. Is that OK or should I be using something different?

--Thanks again for taking the time to read this.

 

[Gordo]

The only thing that jumps out is the unbanded side of the 5817 diode. That solder joint looks a bit iffy on both the back and front but it might just be the camera angle. Also is R3 5K1 or is it 51K. Again, might just be the color in the picture. Other than that you had the foresight to socket the transistor, which will make troubleshooting the BS170 much easier.

 

[jpants]

whoa, thanks for such a quick reply,

R3 is def 5K1, I'll try reflowing the diode after dinner

 

[Barry]

I'd reflow all those solder joints that look more like a blob than a cone

 

[Chuck D. Bones]

Q1 pinout:


Make sure that Q1's leads are snug in the socket. The socket you have is for ICs. Transistor leads don't fit properly. Bend the leads slightly at the ends so they'll fit snugly.

MOSFETs are very sensitive to static electricity. When it's installed on the board, D1 protects it. When it's not in the board, it's vulnerable to damage. Don't work on this on a dry day.

Do not try to measure the DC voltage at Q1's gate, you won't get a good reading because your meter will load down the circuit. Make all voltage measurements with the black (-) DVM lead connected to ground. The voltages you get on Drain and Source will vary depending on the GAIN setting. Make measurements of the Drain voltage with GAIN at min and at max. and report them here.

With GAIN at minimum. the signal levels that you hear with an audio probe at Q1's G, S or D should all be about the same. Tell us how you built your audio probe and what it's plugged into.

A note on flux: Solder has flux in it. If you don't know how to use flux, then don't use it. Just rely on the flux in the solder, you'll be fine. I use flux very sparingly. Most of my solder joints do not have additional flux beyond what's already in the solder.

Good luck.

 

[jpants]

I reflowed the connections again and cleaned off more of the flux residue. I'm still in the same boat with the output. Here's the pics of the current state. (still a couple bad joints now that i'm looking at the pictures)


Audio probe is a mono jack with ground connected to a wire and aligator clip connected to ground and the tip lead connected to a 100nf cap and a solid wire for probing. I plug that jack into the 1/4" cable going to the amp while the input of the ciruit is connected to a signal coming from a guitar and loop pedal.

Audio probe results are giving about the same volume on all pins of Q1 with gain at min, but with gain at max there is no signal on any of them. I trimmed the leads and pushed it in as far as it seems to go but I haven't tried any of my other bs170s yet.

Drain Voltage with black lead of multimeter connected to ground, red lead connected to Q1 drain:
gain at max (all the way clockwise): 0v
gain at min (all the way ccw): -3.85v
volume pot is maxed for both of these.

 

[Chuck D. Bones]

I think you have the meter hooked up wrong. You should only have positive voltage in that circuit. The Drain lead is the one closest to the green wire in the picture above. Ground is the green wire. Is the black meter lead plugged into the COM or - jack on the meter?

 

[jpants]

So maybe my multimeter isn't very good it's definitely a cheapo, but I've never really been able to get any good readings with the black lead connected to COM or I'm doing it wrong.

If Drain is the one on the far right when viewing component side of board top down with LED up, then the readings are as follows with black lead on ground and red lead on drain of Q1
gain knob at MIN: -5.45
gain knob at MAx: -2.86
with multimeter hooked up like this pic:


When the black lead is connected to COM like the following pic, I get 0v from Drain to ground with gain knob at both max and min




I'm done for the night, I really appreciate all the feedback

 

[Chuck D. Bones]

Yeah, of the three possible hookups, you pick the two wrong ones. The hookup in the first pic reverses the voltage at the meter. The hookup in the second pic could have damaged something. You probably got away with it this time.

Black probe goes into the COM hole. Red probe goes into VΩmA. Do yourself and your meter a favor and NEVER use the 10ADC hole. Also, do not switch to the DCA range on the dial, not even for a moment, until you understand what that range does. Otherwise you risk blowing your meter up.

Now you can go back and repeat the measurements.

 

[jpants]

Thanks for clarifying the meter blunder.
so now I'm reading 2.86v with gain at max and 5.46 with gain at min

 

[Chuck D. Bones]

Those voltages look a little low, but not out of spec. If you have another BS170, try swapping it out.

 

[jpants]

So if I am understanding correctly, I'd ideally be getting the full 3.0v dc listed as Vgs(th) on the spec sheet when gain is maxed. What value am I aiming for when gain is min?

I'm going to a concert tonight so I won't get to try this until tomorrow evening. I do have 3 other BS170s that I could try, I also have a couple of 2N7000s if that will work, although what I've read seems to indicate that despite the similar gate threshhold voltage that it will behave/sound a little different.

I'm trying to research as much about mosfets and simple circuits like this one as I can. In addition to a few random videos I've mostly been reading here and here. If anyone has any good additional reading recommendations I'd greatly appreciate it as I'd like to learn as much as possible about whats really going on.

Thanks again for everyone's help so far, I really wasn't expecting such quick and helpful responses. I feel like I'm making progess even though the effect isn't working yet. I'll update tomorrow with the results of swapping transistors.

 

[Nostradoomus]

2n7000 should work the same, just opposite pinout so flip em around.

 

[Chuck D. Bones]

BS170 gate threshold voltage is somewhere between 0.8V and 3.0V. Varies from device to device. When GAIN is maxed, the drain voltage will be 2x the gate voltage because R1 & R2 make a 2:1 voltage divider. Because the drain current is low in this circuit, Q1 will be operating just above the threshold voltage. The means the drain voltage can be anywhere between 1.6V and 6.0V when GAIN is at max. I suspect the problem is not with Q1, it's most likely a bad solder joint somewhere.

 

[jpants]

The circuit is now working. I tried a quick bs170 swap out and had the same results. Then I reflowed some trouble joints and at first it wasn't working with the original bs170 so I was about to stop for the night but decided to try yet a 3rd bs170 and it suddenly came to life.

This one tests much higher. 4.43v with Gain Max and 7.58v with gain Min.

Now I'm kinda tempted to try all the bs170s and 2n7000s that I have just to see which work and what those readings are before I box it up.

Thanks again for all the help

 

[jpants]

update:
I tried all my bs170s, they all work except the original one. all the working ones give a drain voltage of about 4.3-4.5v with gain at max
The original, now reads low, around 0.7v.

Bad soldering was definitely the culprit because swapping transistors did not work prior to the reflow.

I'm assuming I damaged the original somewhere along the way. So I'm gonna box this up with one of the working ones and see what happens.

also again thank you all for being so helpful and responsive.