Footswitch Cut Sound

Forum dedicated to helping people debug and troubleshoot non-functional pedals or builds. Please use an clear and informative title, indicating circuit and basic problem. Don't forget to mark the issue as fixed if this is the case.
Post Reply
User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

Hello!
I just finished to build an octavia (exactly an octahive) that i moded a bit to add a 1 band EQ.
The sound is great, everything works fine.
But! [smilie=madflame.gif]

I installed it in a box with a standard 3PDT switch.
And when i switch from clean to fuzz every 2 or 3 sec everything is fine.
But if i wait 10 sec or more se sound cuts and comes back distorded and after it's back to normal (this happen when i go from clean to fuzz and also from fuzz to clean)

(i tried with 2 other footswitches, i tried to change the 4.7Mohm to a 1Mohm, and all the wiring is good)

Anyone as an idea? :)
Attachments
octa.jpg

User avatar
stolen
Breadboard Brother
Information
Posts: 129
Joined: 04 Mar 2021, 09:54
Has thanked: 44 times
Been thanked: 55 times

Post by stolen »

Hey,

first of all, nice mod :D! It does sound like you're dealing with bias shifts here. Since the device is working fine on its own, and it happens when switching from fuzz to clean too, it's likely that it occurs at the input/output stages of the surrounding devices. While your schematic suggests that input and output are properly DC-blocked, it would be very helpful that you could verify with a multimeter that the bias voltages at the input and output are indeed at 0V with no input signal applied (but the pedal being powered). If this checks out without input and output connected, you might wanna repeat that test while having it engaged and hooked up to the rig where you observed the effect.

If that is not the case, we'd ask you to collect a bit more data: This type of octaver is a rectifier, and therefore by definition has variable internal bias depending on input volume changes. Does the cutting out also happen when you have fuzz engaged, but are not playing for like 20 seconds, then turn it off into true bypass and immediately start playing?

All the best,
stolen

User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

Thank for your answer.

I feel really really stupid.
But thank to you i found the problem
(I didn't show you the full schematic, cause i bypassed a part of it (the 2nd part of the Op amp that i wanted to use to raise the volume, but it wasn't needed))
however the 7 th leg of the Op amp was connected. And that was the issue.

Anyway, thank a lot for your fast and usefull answer :D
Best
Attachments
octa 2.jpg

User avatar
stolen
Breadboard Brother
Information
Posts: 129
Joined: 04 Mar 2021, 09:54
Has thanked: 44 times
Been thanked: 55 times

Post by stolen »

Oh fair, that explains things :D. Glad it worked out!

EDIT: Didn't properly look at it and got curious again; you might wanna disconnect R18 too to properly remove DC :3

User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

Hahaha yes i didn t solder the R18 so i m good :)
I do have a last question.
So, I cloned the octahive by beetronic that
I opened up.
I m not too far, but I m not there yet. The original has way way more output and gain.
I checked everything like 10 times.
I checked the transistors of the original one. And I found something weird.

On Q3 (with a diode tester) I have 0.8 between the emetter and the base.
I found the same between the collector and the base. (Like on all the other 5088 that I have)
But I found 1.4 between the base and the collector (and nothing on all the others).

I don't know what I have to conclude about that haha .

User avatar
stolen
Breadboard Brother
Information
Posts: 129
Joined: 04 Mar 2021, 09:54
Has thanked: 44 times
Been thanked: 55 times

Post by stolen »

Is it the circuit from above, minus the midboost? Transistor variance is a thing, but we'd be surprised if beetronics matched them, and even more surprised if it made such a huge difference. Are you sure about the model numbers for the transistors and the transformer? In the above schematic, Q1 appears to be in reverse active mode, which has a very decreased current amplification. Maybe collector and emitter are swapped?

All the best,
stolen

User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

Yes , same circuit but the output is at the volume knob. (And it s not connected to anything else, I learned from my previous mistake.)
Yes Q1 is swaped I released it later. it made a difference but it wasn t a huge one.
And yes I m sure about the transformer and the transistors. Cause there are all clearly marked.

User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

(Same pots as well)

User avatar
stolen
Breadboard Brother
Information
Posts: 129
Joined: 04 Mar 2021, 09:54
Has thanked: 44 times
Been thanked: 55 times

Post by stolen »

In that case we'd advise to take bias voltages of all transistor terminals. Can you quantify how much volume difference there is (for example, at what position of the volume knob the original's volume matches the clone with its volume knob maxed?

User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

Hello, I'm Back full of datas.
About the volume. to match the clone the original one is at about 3/4 (but there are linear pot, so a lot happens in this last 1/4)
and mine is really really midrangy with less gain.

About the transistors


Octahive /// Clone

Q1. E 4.8V /// 4.2V
B. 3.45V /// 3.7V
C 2.82V /// 2.81V

Q2. E 2.25V /// 2.26V
B. 2.82V /// 2.81V
C 8.85V /// 9.41V

Q3 E 1.59V /// 1.60V
B. 2.25V /// 2.26V
C 4.76V /// 4.87V


(I linked as well a picture of the original circuit)
Attachments
IMG_3573-Récupéré.jpg

User avatar
stolen
Breadboard Brother
Information
Posts: 129
Joined: 04 Mar 2021, 09:54
Has thanked: 44 times
Been thanked: 55 times

Post by stolen »

Hi!

Those bias values look good; we'll assume one of the two 2n5087 are a mislabelled in the photo above, mostly because of this very unrelated drawing. Anyways - in your drawing above, the volume pot is behind a 2.2uF decoupling cap, the diodes should definitely have a resistive path to ground there. With the picture you provided, it seems that the parts list of the schematic we linked matches the pedal quite well, so we'll take them to be the same. Can you confirm that the output stage of your build is identical? At least the loss in volume could be well explained by a lacking path to ground, so that could solve one part of the riddle :3

Just as a side note, C10 is flipped and the mid boost stage can load the circuit quite a bit, we'd recommend a buffer there if you want transparency, but we assume you have it disconnected for now.

All the best,
stolen :D

User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

I removed the 2.2u and it helped a lot
As hell as changing the the 220u cap for an other one (I don t know why but it was the good value but didn t work well )
I wired a 0.1u cap after the volume like on the schematic you sent. (It s actually the same on the Octahive)

So know I m good with the volume and the gain. I have less midrange but I miss quite q bit of treble, so the octave effect is t as clear as the original.
(I connected the collecter of Q3 with R20, and now the voltage is closer to the original.)

It s already a big step 😁

User avatar
stolen
Breadboard Brother
Information
Posts: 129
Joined: 04 Mar 2021, 09:54
Has thanked: 44 times
Been thanked: 55 times

Post by stolen »

Glad we got some progress!
Robin28 wrote: 13 Sep 2021, 07:18 As hell as changing the the 220u cap for an other one (I don t know why but it was the good value but didn t work well )
Electrolytes degrade quickly when heated for too long, so if you soldered it with a lot of heat and time that could explain why the 220u cap wasn't bypassing properly.
Robin28 wrote: 13 Sep 2021, 07:18 (I connected the collecter of Q3 with R20, and now the voltage is closer to the original.)
So you shorted R12...? That isn't the best idea - exactly matching bias voltages isn't that important and will always vary a bit in a transistor circuit, and you significantly change gain and standby current by reducing the collector resistor to a mere 47 Ohm. You might want to undo that.

Treble cuts are happening via the miller cap C4 as well as the network around R8/C7. You could double check these if they indeed match the original. You could try using an audio probe to see if the signals at the inputs of the transformer show a similar discrepancy to take that out of the equation. Note that a less pronounced upper octave often hints at more treble content before the rectifier.

User avatar
Robin28
Information
Posts: 25
Joined: 18 Mar 2021, 13:14
Has thanked: 1 time
Been thanked: 4 times

Post by Robin28 »

HI
Sorry about Q3 and R20 i was talking about Q2 (like on the picture).
Is it still a bad idea?
My problème was C4, i didn't have the right value.
they sound really similar now :)

Thanks a lot for your help.
In my previous build i worked a lot with Op amp, but not a lot with transistors.
Do you know any website to learn more about it?
(Like i had no idea that R8 and c7 was a Low pass filter.)

Anyway really really thank your for all your help :)
Attachments
Capture d’écran 2021-09-14 193157.jpg

User avatar
stolen
Breadboard Brother
Information
Posts: 129
Joined: 04 Mar 2021, 09:54
Has thanked: 44 times
Been thanked: 55 times

Post by stolen »

Oh fair, that makes more sense! Yeah just leave it like that, looks fine! You might wanna open a tracing thread so that people who look for the schematic of the octahive can find it more conveniently (or maybe move this one, we're not quite sure - maybe an admin can advise)!

The standard resource would be the good old Art of Electronics, which can be found as a free pdf online with some googling. Older revisions are fine for all the non-market-specific information. Transistor circuits are often a bit more tricky than OPAs, biasing needs more attention, some design parameters have huge variations, and you're dealing with a lot of current sources, so learning about these basics is best done with a proper textbook in our opinion.

For example, the collector of Q1 is a current source output, which with R7 forms something identical to a voltage source in series with a resistor - R8 is very small in comparison, so mostly that virtual series resistor forms a low pass with C7. The whole circuit is biased with a feedback network that includes all transistors and we'd to write down some linear equations to figure out the bias points; this knowledge is not easily compressed into a forum post. We tried that once with little clarity, it's much easier to understand if a skilled author explains it with some proper illustrations.

Hope you have fun with the pedal! And if it cheers you up, we have a friend who wanted one of these themselves and is now set out to build one based on your trace from what we've been told :D.

All the best,
stolen

Post Reply