MusicPCB Echobase build: No sound

[naught101]

Hey all, I bought the MusicPCB Echobase board, and build it reasonably closely, and I'm getting no sound, just lots of noise. Details:


Name of project: Echo Base Delay
Source: PCB and plans from http://www.musicpcb.com/pcbs/echo-base-delay/
Closeness of layout: bought PCB - no mods.
Part substitutions:
- Some of the odd resistors I didn't have (20k, 50k, 680k, 240k), so I made up some by adding 2 lower values in parallel. I checked them all, they're all within 5%
- Some of the caps I didn't have, so I used the next size up (18nf for 15nf, 120 for 100)
- Didn't have 2N5089, so used BC549B - same pinout (as far as I can tell from datasheets), lower CE voltage (30 not 45), lower collector current (100, not 200mA)
- Didn't have BC560, so used BC559 - lower CE voltage (30 not 45), pretty much the same otherwise
- No tails switch, no LFO switch
- Not sure about directionality of pots - centre to centre, everything else consistent. (I guess if I got this wrong, the pots would just be backwards?)

Build Quaity:
- My soldering is generally fine, I think.
    - I ripped the pad off the +9v through hole on the top, but I scraped some of the solder mask off the lead coming off it, and soldered on to that, and it seemed to work - LFO light works.
- I assume that the in/out jack are grounded to ground (this is the first pedal I've built, and it's not marked on the schematic)

Negative ground circuit.

What works: LF light flashes, LFO rate knob works.
What doesn't work:
- No sound, neither with "bypass" switch on or off.
- Quite a lot of noise (more than an unplugged lead).
- amp pops and noise changes tone when I touch the guitar strings.
- I have tested the guitar, amp, and both leads.

I'm using a wall-wart. Voltave is ~9.2 unplugged, and ~9.1 plugged in.

Voltages (no input or output, directions relative to PCB in PDF):
- Top left TL072: all ~5v except bottom left (0V), and top right (9V)
- Right TL072: same except bottom left (0.8V), top right (0.83V), right second from top: moving between 0.2 and 7V - LFO
- 4066: left, top to bottom: 4.6, 4.6, 1.3, 3.3 (or 1.3), 0, 3.1, 0; right, top to bottom: 9, 8.9, 8.9, 4.6, 4.6, 2.4, 2.4.
- PT2399 : L, TtoB: 5, 2.5, 0, 0, 3.5, 2.5, 1, 1; Right TtoB: all ~2.45V
- BC549 (2N5089): E:0, B: 0.6, C: 6.5
- BC559 (BC560): E: 0.6, B: 0.05, C: 0
- 7805: 5, 0, 9
- 1N4148 diodes: both read 0V on both sides.
- 1N4001: L: 0, R: 9V

These voltages were all taken with knobs in about the middle of the range, but they may have moved a little. Let me know if I need to provide any more exact info.

I've checked a couple of other echobase threads on this forum, but I can't find anything relevant to this problem.

[naught101]

The 2N5089 replacement suggestion was from here: http://www.diystompboxes.com/smfforum/index.php?topic=44348.0

Mine is not a BC549C, but BC549B

[Fender3D]

2N5089 and BCs haven't the same pinout, they have E, B, and C pins though...

[naught101]

2N5089 and BCs haven't the same pinout, they have E, B, and C pins though...

Hrm... I was having a hard time reading the 2N5089 datasheet (http://datasheet.octopart.com/2N5089-NTE-Electronics-datasheet-41970.pdf), is that a flat-side front view? The orthogonal plan view is maybe throwing me off. I will try replacing it the other way and see if that works.


Build photos, while I'm at it.: http://imgur.com/a/cZuH8

[naught101]

Ok, I tried switching the BC549 around, and still no joy. Higher pitched noise, otherwise the same response. Perhaps the BC549 is damaged?

[Granny Gremlin]

Check U3 (TL072).  Half of it is the input buffer and the other half is the output.  If no sound that's the first place I'd look (in fact I had an issue there with one of mine).  If you have noise but no signal I'm thinking input (noise change when you touch the strings could be due to ground not hot signal getting thu the buffer)... which reminds me - check your grounds - you sounded unsure of how that should go - common practice is to ground input to the PCB or DC jack ground and output is grounded by the chassis to input (assuming not powdercoated; if coated, sand away the finish on the inside where the jacks are mounted; check for continuity with a meter).  Before I boxed one of mine up I was testing it and forgot to connect the ground with an alligator clip from the input jack to the output - I had the exact issue you describe (no signal but plenty of noise that can be affected by the knobs); clipped on the ground and fine.

Also take a look at the (very long, but very helpful) Echo Base build/help thread.  Taylor will find you there too if you post in it:

http://www.diystompboxes.com/smfforum/index.php?topic=84109.0

[slacker]

^^^ what he said, especially about the grounds. If you're getting no sound and it's not a ground or wiring problem then the problem is somewhere around the opamp top left on the board. You should be getting dry signal in either setting of the bypass, any problem with the transistors wouldn't affect this.
If the LED turns on and off with the bypass then that transistor should be correct.
The way you have it now, with no tails switch the delay will always be on no matter what position the bypass is in. You need to connect the middle tails switch pad to one of the outer ones depending on whether you want tails or not, sorry I can't remember off hand which side is which.

[naught101]

which reminds me - check your grounds - you sounded unsure of how that should go - common practice is to ground input to the PCB or DC jack ground and output is grounded by the chassis to input (assuming not powdercoated; if coated, sand away the finish on the inside where the jacks are mounted; check for continuity with a meter).  Before I boxed one of mine up I was testing it and forgot to connect the ground with an alligator clip from the input jack to the output - I had the exact issue you describe (no signal but plenty of noise that can be affected by the knobs); clipped on the ground and fine.

Yeah, both input and output are connected directly to DC ground (the three wires heathshrinked left of the LED in the build pictures). I'll double check continuity and than TL072 when I get home tomorrow, and then have a browse through that thread.

[naught101]

The way you have it now, with no tails switch the delay will always be on no matter what position the bypass is in. You need to connect the middle tails switch pad to one of the outer ones depending on whether you want tails or not, sorry I can't remember off hand which side is which.

Ah, ok, good to know. Maybe I will wire a switch in there tomorrow then.

[naught101]

Checked with the case grounded (wasn't before, but both jacks were grounded to DC ground) - no change.

Checked with bypass switch, no change. (leaving it out for now, 'cause I don't care about bypass until I get it working).

If you turn the feedback up, it appears to feedback on the noise (self-oscillation-ish, controllable with the time knob), so I guess the PT2399 stage is ok.

I tried replacing the input TL072 (the top left one). The noise is a bit different, but the overall behaviour is similar - no sound but noise, some change in the noise when I touch the guitar/ground.

I'm not getting readings across the two 1M resistors (no connection), but I think that's 'cause my multimeter is shit (it doesn't read other 1M resistors either).

[Taylor]

Checked with the case grounded (wasn't before, but both jacks were grounded to DC ground) - no change.

Checked with bypass switch, no change. (leaving it out for now, 'cause I don't care about bypass until I get it working).

If you turn the feedback up, it appears to feedback on the noise (self-oscillation-ish, controllable with the time knob), so I guess the PT2399 stage is ok.

I tried replacing the input TL072 (the top left one). The noise is a bit different, but the overall behaviour is similar - no sound but noise, some change in the noise when I touch the guitar/ground.

I'm not getting readings across the two 1M resistors (no connection), but I think that's 'cause my multimeter is shit (it doesn't read other 1M resistors either).

You often won't be able to get any useful readings from resistors or caps in-circuit, because the other parts to which they're connected are forming various networks together. So that's no big deal.

It actually sounds like almost everything's working - the LFO is oscillating, you're getting delayed noise out of the delay chip and through the output stage. As the other have said the issue is going to be around that input opamp, since your guitar signal is neither getting to the delay chip nor through the clean blend path. If you haven't already, build up a little audio probe, and with some kind of continuous audio going into the input, probe pin 1  of the upper left TL072. (Here's how IC pins are numbered, BTW.) See if you can hear your audio at pins 1 or 3.

[naught101]

It actually sounds like almost everything's working - the LFO is oscillating, you're getting delayed noise out of the delay chip and through the output stage. As the other have said the issue is going to be around that input opamp, since your guitar signal is neither getting to the delay chip nor through the clean blend path. If you haven't already, build up a little audio probe, and with some kind of continuous audio going into the input, probe pin 1  of the upper left TL072. (Here's how IC pins are numbered, BTW.) See if you can hear your audio at pins 1 or 3.

OK, with the audio probe, I get audio at pin 3 when the power is not plugged in, and nothing at the other pins. With the power plugged in, On the TL072 I get:
1, 2 and 3: Massive noise, plus some audio signal
4: silence
5: silence
6: massive noise + delayed audio signal
7: quieter, very high pitched noise

Sounds kinda like a short, I guess...

[slacker]

To make it easier to debug I would remove the cd4066 and the pt2399. This leaves just the dry signal so any problems with the delayed signal won't confuse things. See if doing this changes the results of your audio probing.

[naught101]

To make it easier to debug I would remove the cd4066 and the pt2399. This leaves just the dry signal so any problems with the delayed signal won't confuse things. See if doing this changes the results of your audio probing.

OK, I took them out. Doesn't appear to change much, other than the obvious (no delay on pin 7) - so I still get:
1,2,3: Very noisy, some audio signal.
4,5,6: silence
7: similar to 1,2,3.
8: high pitched whine, no audio.

My audio probe is also pretty noisy when not connected, and even more so when my hand is near/on the capacitor. Is that normal?

[slacker]

Have you tried swapping the TL072 for another one, just in case it's faulty.

[Granny Gremlin]

I think he said that he did.  What I would do is pull the 72 and check that the socket is in properly (test continuity between the pin sockets on top and the pads below on the solder side of the PCB - put your probe on the solder mask next to the pad, apply a little downward pressure and slide it towards the pad until it just touches - even better to test between the pin socket and whatever downstream component's pad if you can follow the traces).

[garcho]

Some of the odd resistors I didn't have (20k, 50k, 680k, 240k), so I made up some by adding 2 lower values in parallel. I checked them all, they're all within 5%

do you mean 2 lower values in series? Or 2 higher values in parallel? 2 lower values in parallel gives you something way lower than the original value called for.

[naught101]

Some of the odd resistors I didn't have (20k, 50k, 680k, 240k), so I made up some by adding 2 lower values in parallel. I checked them all, they're all within 5%

do you mean 2 lower values in series? Or 2 higher values in parallel? 2 lower values in parallel gives you something way lower than the original value called for.

Series - I'm ok with basic electronics, I just haven't done much with audio before  .

As I said, I checked them all. Actually, I remembered that I used 22ks for the 20ks, instead of trying to make up custom ones. The other 3 are all pretty accurate.

[naught101]

Build Quaity:
- My soldering is generally fine, I think.

*sigh*... U3 pin 4 had a cold solder joint, which was even visually obvious after I noticed it via testing. Sorry for the run around, and thanks for all the help, I think I learned a few things about debugging.

The pedal sounds pretty good now. I'm not getting much noise at all, and I suspect what I am getting comes mostly from the guitar and lead. I can easily get 20 clean repeats. After about 2 O'clock on the feedback knob, I start getting self oscillation, and some crazy cool noises as a result.

One thing that I'm not sure about is that if the guitar volume is up too high (more than 3/4, perhaps), then the first echo is fairly noisy. I guess it's maxing out the op-amp? Or part of the PT2399? Is that normal? I can totally live with it if so, but it'd be cool if it were fixable.

[garcho]

the PT2399 distorts easily; low headroom. look at the build document for the "humbucker lift" mod.