Hey everyone first time poster here,
I recently put together a GGG EA trem and I am getting no signal through the circuit at all. I am very new to pedal building, this is only my second project after a full-pedal version of the byoc confidence boost. I am getting signal in bypass. I did the flashing led mod and the led comes on and flashes with the speed knob. I noticed when I turned the volume on the pedal all the way up I could hear a little bit of noise coming out and that noise responded to the depth and speed knobs. I have a multimeter and an audio probe, but I don't even know where to start with them as I've never had to debug anything before. I do know that signal is at least getting to the input of the board, but I don't know where to test from there.
Thanks for any help and advice you have!
Welcome to the forum, DClubb! No worries, in a little while you'll master debugging. It's normal for this to happen, we all make mistakes and so on.
1) http://www.generalguitargadgets.com/pdf/ggg_eat_sc_improved.pdf (http://www.generalguitargadgets.com/pdf/ggg_eat_sc_improved.pdf)
2) Audio probe where R15 and R16 meet, while you input some signal (someone playing guitar, a recording, etc). Any sound? Make sure you have a battery in, the jacks are plugged in, and activate the switch, of course! (sorry if I am repeating what you've already done, working fast here, dinner time)
If you have no signal, the problem is likely your switch wiring, or jack ground switching. Battery dead? LOL Do you see 9V at the 'top' of R15? Often, I will 'inject' audio right into the circuit before adding jacks or switch...at left side of C8. Then if I get it at the output, I know the circuit works. R.G. Keen designed a "quick and dirty oscillator' (net search) to do just this, future project for you...
If you do get signal at that point, test at C1, R12....find where it stops and report back. That is the audio path, the other stuff is the LFO that trems it. Those test points will tell us which stage isn't working (hopefully). Usually there is an error with a resistor value (placed wrong), a solder bridge, cold solder joint, missing ground, a Q put in with the pinout the wrong way around and so on. Go back over all your work, even w/a magnifying glass! You'd be surprised....
If above didn't work, visit the debugging page and provide the info requested best you can (back here, below this).
https://www.diystompboxes.com/smfforum/index.php?topic=29816.0 (https://www.diystompboxes.com/smfforum/index.php?topic=29816.0)
Thanks for the help!
I do have signal where R15 and R16 meet but I've got nothing at C1. I'm pretty sure the pinouts on my Qs are right because the pcb from ggg has the flat side marked. I would call my soldering skills sub-par at best, so I'm imagining my problem lies there, but to my untrained eye, everything seems to look alright.
Forgive me if this is a dumb question, but I have very little knowledge about what the various components in a circuit actually do. Should I have audio signal at multiple pins in Q4 (2N5457)? Because I have signal at pin 3, which I think is gate (I do not know what this means)? But not at the other 2. I don't know if this helps.
Is your JFET a 2N5457? Can you read the part number on it?
Ok, step by step. If you have signal at R15/R16, that is the same as at the gate of Q4 - the input. On the schematic, you can see they all share that node, all should be connected. That JFET is acting as a buffer; you should read ~9V on the drain, and have signal on the other side, the source, which is connected to C9 and C1. Q1 is similar, but is a gain stage because the output is taken from the drain rather than the source (don't worry about that right now) - does double duty, the source is 'tickled' by the LFO to do the tremolo action.
Because you have no signal at C1 (do check back with audio probe at C9 and the leg of the JFET it is attached to, please, JUST in case something is wrong w/those caps), suggests that stage is not functioning. Need voltages on the 3 legs, labeled "D S G" if you can. Source is the one connected to 10k/C9....gate is the input R15/16, and the drain is the leg simply going to 9V. Check that the values of R15 and R16 are what they are supposed to be, and that the source has that 10k resistor there to ground. When working, that stage should just give you a near approximation of your input signal - not louder. It is a current amplifier, not voltage (again, don't worry 'bout that now). Sound should be nice and clear tho, just like when you strum at the input.
The game is to make sure each stage operates, one by one. This is not a complex thing, tho it looks that way at first. (and my descriptions may make it seem like it is!) Voltages of battery and Q4, please :)
For reference, this is the signal path. Sorry the pic is so big, LOL.
(https://i.imgur.com/vxqB5sG.jpg)
Pardon me Sir, but your traceability path tends to be quite deceptive.. :icon_redface:
(from original signal path point of view..)
P.S.
Of course, your signal path draw is exactly what OP must check for..!! :icon_wink:
Quote from: Dclubb83 on July 16, 2019, 07:06:27 PM
I'm pretty sure the pinouts on my Qs are right because the pcb from ggg has the flat side marked.
we do not allow this assumption around here, and
welcome to around here. because - if you sub a BC548 for a 2N5088 or 9, the pinouts are opposite. if you sub a 2N5457 with a 2N5952 or a BF245B, or worse, a BF244 or 2SK30A, your gate pin has gone out the window. and then there's the 2N7000 and the BS170, opposites again. these are all common swaps, we see them weekly.
so, check the part numbers
on your transistors, look up the datasheets, have a look at what is there. you need to get familiar with them sooner or later.
Wow thanks guys!
Ok... so I am indeed using a 2N5457 as Q4 and I checked the pinouts and the pcb orientation was correct. I am getting signal at gate but not at source. The voltages are
D - 9.3v
S - 0
G - 3.11v
I'm assuming that 0 has something to do with it...
Also I am using a power supply and not a battery as a source and the output of that supply is 9.3v
Not cool. I threw that stage on the breadboard and get 9V, 5V, 4V respectively. Your gate is close, anyway. Can you test that each leg is connected where it should be (continuity)? I would suspect it's toasted unless you have a bad joint, or the source is connected right to ground somehow.
Can you post a pic of the front and back of the board showing the FET? Duck LOVES that, he has good eyes to spot errors!! :)
Apparently my multimeter doesn't have a continuity mode? So I can't check that, unless there's some other way to do it. Here's the board
(https://i.postimg.cc/jn4Cq0L0/20190717-144121.jpg) (https://postimg.cc/jn4Cq0L0)
(https://i.postimg.cc/G80pkhDn/20190717-144201.jpg) (https://postimg.cc/G80pkhDn)
Look between the 2 big holes, Clubb...is that a wire tip poking thru w/no solder? Did the trace lift? If it did (nothing to solder to)...you can scrape, gently, the green crap away to expose the PCB trace copper and use a piece of resistor lead, etc, to bridge it. If I'm seeing things, pls disregard.
To use your meter for continuity, set to "ohm" (omega symbol) Touch probes together....you should get zero, or very very low, like .1 ohm. You can use this to see if points on the board are connected...you'll read that magnitude of number if they are! If not, you'll get oddball things...100s of ohms, 1,000s etc. You should read 10,000 ohms from the JFET source to ground, for example (check that; pin 2 on the thing, to ground).
That wire tip is the wire leading to the led. It is soldered on the top side because apparently it's original solder connection was bad and it fell out after I boxed the effect when it wasn't working and I gave up on it for a while, so I just re-soldered from the top. The connection seems to be good, because the led works
When I tested the source to ground, I got exactly 10k ohms.
Just to "isolate" Q4 issue (and verify rest circuit good working..), bypass Q4 by shorting with a wire Input & C9 positive leg..
(you will notice a significant volume drop, due to R17 loading, but the rest of the circuit should work..)
I think that FET is dead. Since you have 10K between it and ground, the source should be 'elevated', and not read zero. Like Antonis suggests, bypassing the stage to see if anything down the line works is a good idea.
Next up is probably removing Q4 carefully, careful not to ruin the pads, and replacing it. They CAN be sensitive to heat, so work quickly. I try to have the iron on a leg for no more than a "1, 2" count. Then a pause for it to cool before the next leg. I also leave some space above between the jfet/bjt/mosfet and the PCB. You could use a heat sink clamp, too.
QuoteWhen I tested the source to ground, I got exactly 10k ohms.
do this measure again,
from the leg of the jfet to the circuit ground. please. [my eyes are sore, Sir Mike, but] can we have a pic of your socket wirings, please?
Quote from: duck_arse on July 18, 2019, 10:21:36 AM
QuoteWhen I tested the source to ground, I got exactly 10k ohms.
do this measure again, from the leg of the jfet to the circuit ground. please. [my eyes are sore, Sir Mike, but] can we have a pic of your socket wirings, please?
He could have an issue with that ground 'area' on the PCB not being connected to CIRCUIT GND, absolutely. I'd put that black probe all the way over on the battery minus terminal (being sure my plugs are inserted.....and posting a pic of them for Duck). Since they switch ground, a problem there CAN easily make the thing not work.
Ok so I ran a wire from the input terminal on the footswitch to the positive lug of C9 and BOOM audio all the way though, all of the pots were working as they should. But I didn't notice a volume drop at all. Honestly, the whole pedal seemed to be working as I would've thought it would.
Oh and when I tested the source pin of Q4 to ground I had the output jack screwed in to an enclosure and I just kept the black probe on the enclosure and got the 10k reading.
So now I'm curious as to what that Q4 stage does if bypassing it doesn't seem to have a negative effect on the sound. (I could be wrong about that because my input signal was just a song playing through my phone and the amp I had my audio probe plugged into is a Danelectro Honeytone lol) but the volume pot definitely boosted the signal above dry levels, and the depth and speed pots controlled exactly those things.
Because the signal is taken from the source for output (rather than the drain), Q4 is set up as a buffer. A buffer is a current (rather than voltage) amplifier. That means it can drive lower impedance items connected to it - it also presents a high input impedance for whatever its source is (your guitar). This preserves high frequencies, among other things.
More here:
http://www.muzique.com/lab/buffers.htm (http://www.muzique.com/lab/buffers.htm)
The 'noticeable difference' with a buffer is usually described as 'better high freq. response', or the preservation of 'sparkle'. I guess if Q1's input impedance is rather low (tho it's not REMARKABLY so), they put in the Q4 stage to isolate the tremming (?) stages, to keep them from loading the input signal. I am surmising this.
Q4 appears to be dead. But you know the thing will most likely work when that is taken care of! :)
Quote from: Dclubb83 on July 18, 2019, 07:27:18 PM
Ok so I ran a wire from the input terminal on the footswitch to the positive lug of C9 and BOOM audio all the way though, all of the pots were working as they should. But I didn't notice a volume drop at all.
So now I'm curious as to what that Q4 stage does if bypassing it doesn't seem to have a negative effect on the sound
What Sir Mike already told you.. :icon_wink:
In case of no volume drop at all, there should exist 3 cases:
1. Ultra low guitar (or other source) output impedance (less than 10% of R7..)
2. NO connection between C9 positive leg & R7 upper leg..
3. NO connection between R7 lower leg & GND..
3
rd case is highly suspected in conjunction with Q4 "dead" symptoms.. :icon_wink:
Quote from: GibsonGM on July 18, 2019, 02:55:06 PM
Quote from: duck_arse on July 18, 2019, 10:21:36 AM
QuoteWhen I tested the source to ground, I got exactly 10k ohms.
do this measure again, from the leg of the jfet to the circuit ground. please. [my eyes are sore, Sir Mike, but] can we have a pic of your socket wirings, please?
He could have an issue with that ground 'area' on the PCB not being connected to CIRCUIT GND, absolutely. I'd put that black probe all the way over on the battery minus terminal (being sure my plugs are inserted.....and posting a pic of them for Duck). Since they switch ground, a problem there CAN easily make the thing not work.
I was going to let it slide, but that 10k, exactly. we know the resistor is 10k, because
we meter tested all our resistors before we started soldering, obviously. and because when tested in circuit, they most likely will read "off" - parallel components, semiconductor junctions, discharging caps, etc.
so I wanted to test that the jfet leg actually connected to a resistor that was actually connecting to a ground, not just a 10k resistor possibly connecting to nothing at one end, say. which might then give us a resistance reading, in circuit, of 10k, instead of the maybe less we usually see. see?
I hear you, Duck, and totally agree. The caps that are also attached absolutely can (and should!) skew the reading. Working with OP re. continuity testing...I'd actually 'beep' out source to the 10k resistor, then beep its other leg to ground to be sure. If that resistor is of another value, it sure could drag down Q4 enough to cause it to not work!
Clubb, are you following that? You should identify the source leg, make SURE it is going to the 10k, and make sure 10k is going to ground - with good solder joints...if the 10k is NOT 10k, you may replace Q4 only to find it still doesn't work...
Ok I'm not gonna lie. You guys are kinda losing me and a lot of this is going pretty far over my head. I'm trying to keep up, but I don't think I'm getting everything.
When you build a kit, sometimes it's easy to accidentally swap resistors, and put the wrong one some place! So, if the source resistor at Q4 isn't 10k, it could cause problems (like, no sound). Duck wants you to make SURE you actually have a 10k there. Measuring source to ground might lie, though....capacitors also show up having resistance when you put a meter in a spot that is connected to them (it increases over time as you watch, generally).
That's why I didn't get too specific on you, ha ha. You could unsolder the resistor attached to the source, AT the source, then read that lifted end to ground, see if it's 10k. I hate the blue resistors, I can't read them for crap (old school I guess). That would separate it from C9, so you could be sure of the reading.
I still lean towards Q4 being dead, but that's a gut thing. You will 'get it' over time, LOL. Debug enough stuff, those 'tricks' will come to you! I'm trying to find a way for you to fix this easily...if you have another 5457, I'd say take out Q4, check that resistor, pop in the new one, done. But then you might mess up the pads on the PCB, so Duck is trying to see if that is REALLY necessary, or if it's just that maybe that resistor is the wrong value.
I checked that resistor, it is indeed 10k. Seeing as how this is only my second project and both have basically been kits, I do not have any spare parts. I am going to stock up on various components from Tayda soon and I'll definitely be sure there are some 2n5457's on there and I'll swap one out and hopefully that'll do it.
In the mean time I suppose I'll look over all of your guys' comments and try to make head or tails of everything that confused the hell out of me.. which was about 75% of it lol.
Thanks for all the help so far! You guys are great!
Just be sure to come back with more questions, DC! :)
So, did we or didn't verify 10k from JFET Source leg to GND..??? :o
with all due respect, Sir Mike, I humbly submit that I actually wanted the opposite .....
Quote from: GibsonGM on July 19, 2019, 05:32:04 PM
When you build a kit, sometimes it's easy to accidentally swap resistors, and put the wrong one some place! So, if the source resistor at Q4 isn't 10k, it could cause problems (like, no sound). Duck wants you to make SURE you actually have a 10k there. Measuring source to ground might lie, though....capacitors also show up having resistance when you put a meter in a spot that is connected to them (it increases over time as you watch, generally).
we know his 10k is 10k, what I wanted was proof that the jfet was connecting to ground through that 10k. so put the black meter probe on ground somewhere [but not the resistor leg], and the red meter probe on the leg, the actual pin sticking out of the jfet, so we could be sure of all the solder connections and all copper the traces between. and because his 10k read exactly 10k in circuit, it got me thinking about what he was and wasn't measuring.
and apologies to you to, Dclubb83, for talking about you so rudely in the previous para.
Quote from: duck_arse on July 20, 2019, 11:20:09 AM
we know his 10k is 10k, what I wanted was proof that the jfet was connecting to ground through that 10k. so put the black meter probe on ground somewhere [but not the resistor leg], and the red meter probe on the leg, the actual pin sticking out of the jfet, so we could be sure of all the solder connections and all copper the traces between. and because his 10k read exactly 10k in circuit, it got me thinking about what he was and wasn't measuring.
This is what I did the first time, right? I had the output jack attached to the enclosure, I touched the black probe to the enclosure and I touched the red probe to the source leg (pin 2) of Q4 and I got a 10k reading. I thought it was a way of "2 birds"ing it. I could check the connection of pin 2 and also make sure everything was grounded correctly. Sorry if I'm doing a bad job of explaining stuff.
Quote from: duck_arse on July 20, 2019, 11:20:09 AM
with all due respect, Sir Mike, I humbly submit that I actually wanted the opposite .....
Quote from: GibsonGM on July 19, 2019, 05:32:04 PM
When you build a kit, sometimes it's easy to accidentally swap resistors, and put the wrong one some place! So, if the source resistor at Q4 isn't 10k, it could cause problems (like, no sound). Duck wants you to make SURE you actually have a 10k there. Measuring source to ground might lie, though....capacitors also show up having resistance when you put a meter in a spot that is connected to them (it increases over time as you watch, generally).
we know his 10k is 10k, what I wanted was proof that the jfet was connecting to ground through that 10k. so put the black meter probe on ground somewhere [but not the resistor leg], and the red meter probe on the leg, the actual pin sticking out of the jfet, so we could be sure of all the solder connections and all copper the traces between. and because his 10k read exactly 10k in circuit, it got me thinking about what he was and wasn't measuring.
and apologies to you to, Dclubb83, for talking about you so rudely in the previous para.
I thought I asked him to do the same, and he did...I'm trying not to overwhelm the poor guy, he's new :icon_lol:
"To use your meter for continuity, set to "ohm" (omega symbol) Touch probes together....you should get zero, or very very low, like .1 ohm. You can use this to see if points on the board are connected...you'll read that magnitude of number if they are! If not, you'll get oddball things...100s of ohms, 1,000s etc. You should read 10,000 ohms from the JFET source to ground, for example (check that; pin 2 on the thing, to ground)."
excellent. and apologies for harping. so, we can rule out a problem connection there. carry on.
Nah, there's no harping here, buddy! Don't want to overlook one little thing that could be the key, and have him get all frustrated and give up!
Tayda didn't have any 2n5457's in stock.. so I grabbed a 2n5458.. threw in some sockets plugged 'er in and ITS ALIVE. Lovely lovely trem. Now this has only been tested with some audio from my phone through a Danelectro Honeytone, so I haven't actually ran it for real yet but it seems to be doing everything I want it to. You guys are the best! Hopefully I won't have to revive this thread with more problems in the future! Thanks for everything. Time to move on to a tagboard effects layout of a Super Fuzz for a friend. Wish me luck!
luck. do you prefer good or bad?
Haha it seems like I should have specified because I'm sitting here with no signal on this fuzz. I'm gonna see what I can do about debugging this guy myself. The Super Fuzz is one big ol fuzz circuit...