?problem with WH Purple Platypus

[Kalimna]

Greetings folks,
I have a question regarding my build of the Way Huge Purple Platypus, layout from this here site :
http://tagboardeffects.blogspot.co.uk/2017/04/way-huge-purple-platypus.html

In a nutshell, it works but I think something is amiss. I get a very fuzzy signal coming out that is only subtley affected by the controls. After the pedal has been switched on for a short time, there seems to be a sortof noise-gated effect whereby the guitar signal is extremely quiet if played gently and then back to full volume if played harder. But this full volume decays very quickly. It is hard to describe, sorry!

This happens with any combo of NPN/PNP transistors I have tried.

I wonder if you might be able to suggest where I might start troubleshooting? I'll measure some voltages tonight when I'm at home, though I have tried a couple of different CD4049UBE's with no change.
I did wonder if either one of the capacitors might be faulty (all the electrolytics are from a cheap bundle-pack) or the transistors biased poorly through incorrect resistor selection.

Any thoughts?

Cheers,
Adam

P.S. Otherwise my recent additions seem to be doing the job nicely - MB GravityWave (glorious), correctly re-wired the leads to pots on Timmy w/LM833 (again, glorious) and a Marshall Bluesbreaker (why haven't I built this before???)

[Marbles]

Can you specify what transistors you used?

The effect sounds like what you get when you put a transistor in reverse. Since the transistors aren't specified in the layout, that would be my first guess.

Can you post a picture otherwise?

[Kalimna]

Hi Marbles,
I'll get some photo's up tonight along with the voltages, but I think I have the components the correct way around (having followed the schematic and updated layout). However, that assumes I have the correct ECB orientation on the specific transistors I have. I did google the pinouts, but understand these can sometimes differ depending on producer.

The transistor pair currently in the unit are a 2N3904/3906, but I have tried BC550/560, 2N2222/2N2907 and another pair that escapes me right now. All showed similar characteristics. Also a couple of combinations of non-complementary pairs for the hell of it, just to see.

Daft question time - if the transistors are reversed, why would a signal come through at all? Also, do you mean I may have placed the NPN and PNP in their incorrect sockets, or the ECB orientation for each is incorrect.

Many thanks for your help. I *am* slowly learning about the electrickery behind the effects, but it is a long process!

Adam

[Kalimna]

Ok, here's a link to a couple of photo's :







The red tape is covering the LED resistor and also insulating the output jack socket from the enclosure - I drilled the hole a little close to the top.

And some voltage readings (yes, I realise there is a missing transistor in the photo's - it was replaced for the test). The 3 numbers following each testpoint are with both control pots fully off, halfway and fully on.

Supply - 8.94V, Q1 = 2N5087, Q2 = 2N5088

Q1 e - 1.9/2.6/2.6
Q1 b - 6.1/5.4/5.4
Q1 c - 6.7/6/6

Q2 e - 0.76/no change/no change
Q2 b - 1.4/nc/nc
Q2 c - 3.9/nc/nc

CD4049UBE (minimal or no change with pot settings)
1 - 5.7
2 - 2.4
3 - 2.4
4 - 1.1mV
5 - 5.7
6 - 0.5mV
7 - 5.7
8 - 0.2mV
9 - 5.2
10 - 1.4 mV
11 - 5.7
12 - 1.9mV
13 - 2.5mV
14 - 2.4
15 - 2.4
16 - 3mV

I thought the voltages around Q1/Q2 looked off, so I knifed the tracks again and had slightly different readings with collector at around 8V and gradually decreasing as the gain/vol pots turned fully on. base and emitter voltages changing in a simillar fashion.

Thoughts?

Adam

[Marbles]

The are not that high res, so it's kind of hard to zoom in. I myself am not tech savy at all. Like, at all.So what I say can be nonsense, but.

I would say pin 8 of the 4049 should be zero cause it's grounded, but maybe this is acceptable?

There seems to be some soldering on the right side of the 4049 (around the 68nf and 33nf). But you said you knifed everything again, so that might be solved.

On your pics, the D2 and 100k resistor next to it are hidden. Are they actually there?

[Kalimna]

Hi,
Thanks for replying. To answer your questions - sorry about the resolution, when I was uploading the images seemed truncated at the higher res. Here's a link to zoomable images on imgur https://imgur.com/a/5Nw2a

The 100k and D2 are there - just hiding behind some wiring. Also the lower connecting wire of the 100k is in the correct hole - I made a mistake marking out the track breaks on the component side which was rectified for the soldering side.

The not-quite-zero readings had me a little concerned too, but my DMM had a slightly fluctuating reading when open so I am making the assumption that it is OK. However, maybe the meter needs a new battery? It's only a year old though.

I will go over the tracks again, but as you say I had knifed them after the photo's were taken.

Again, your time/help is much appreciated.

Adam

[Marbles]

Thank you!

Your 10uf Elc Cap on the left seems to be upside down. Unless its twisted, it should be ground down instead of up.

[Kalimna]

Aaaaargh. Right that was a silly mistake, thanks for spotting it.
I've reversed the cap, and checked over the remaining components again. Unfortunately no improvement to the sound, so I suspect there's another problem with my build - dodgy soldering/solder bridge I've missed/faulty component.
It's not a very large board, so I'll populate up another one and see how I get on, and if there is any difference there.

One thought that occurred - I recall from the WH Red Llama thread that it is important to use the unbuffered version of the 4049 for the Red Llama. I have used the same IC here, with the 'UBE' suffix. Might it be the buffered version that is necessary?

Again, thanks for your help.

Adam

[Kalimna]

A brief update. Tried reversing the wayward electrolytic - no change. I then decided to start afresh with a new layout. Once wired up - exactly the same problem with fresh components. I tried it outside of the enclosure in case of some grounding issue - no difference. So I am left with a couple of possibilities :
1) The effect just isn't very good
2) The layout here is in error (seems unlikely)
3) I have 2 faulty 4049 batches
4) There's some other constructional error I haven't accounted for (more likely, but I can't think what it is)

Before I consign this to the 'failed experiment' box and re-purpose the enclosure to another Lovepedal clone, does anyone have any other suggestions where I might look for errors?

If it's of any importance, the effect-is-on LED stays on for a few seconds after the power is unplugged (which I'm guessing is the 330uF discharging).

Thanks again,
Adam