I built the Equinox II (http://www.valvewizard.co.uk/equinox.html) and I get sound coming through it, but it sounds identical both on and off. In fact, I found (after trying 6 different PT2399s) when on I can remove the PT2399 and it doesn't change the sound at all.
This makes me suspect something is wrong with the JFET switching. Does that sound about right?
If so, any tips to help understand and debug what might be going on? (i tried swapping out the J112 with no luck).
UPDATE: In case anyone with the same problem stumbles upon this thread... In my case this problem was because the pot was not hooked up. I mistakenly soldered it to the ground pads. The result is a dry signal only, with no reverb mixed in.
(https://i.postimg.cc/k21GcpjF/equinoxschem.jpg) (https://postimg.cc/k21GcpjF)
Short D and S of the FET with a little scrap of wire. Now it is ON.
(https://i.postimg.cc/7fxDb1S5/mark2-42.gif) (https://postimg.cc/7fxDb1S5)
If that works, re-re-check FET pinout, read voltages around FET with switch both ways. If that does not bring up reverb, your problem is elsewhere. A listening probe may be handy.
Thank you. That makes a lot of sense.
I tried it and unfortunately I still get a sound with no apparent reverb. I can even take the PT2399s out and it doesn't affect the sound. Removing the jumper has no impact either. Mix pot doesn't seem to affect it.
Any other tips or suggestions are very welcome. Otherwise I guess I'll try to use a probe to figure out what part of the signal path isn't being utilized.
QuoteOtherwise I guess I'll try to use a probe to figure out what part of the signal path isn't being utilized.
Actually this isn't going as well as other easier debugging I've done (e.g. where signal was being grounded, or something). Because the signal is getting to the output, and because I don't fully understand yet how the pt2399s are supposed to be adding the delays, it's still a bit over my head.
I'd love any tips or ideas, otherwise I'll keep studying and/or shelf this one for now.
Is your TL072 socketed also? If you still get sound with that removed, you have some sort of switching problem that's bypassing the effect circuitry.
Pictures might help.
Thanks. The TL072 is socketed and it doesn't make any sound when I pull it out.
I added a couple of pics in case it helps anyone. I may try again in the daylight since my office lighting is terrible and the solder joints seem a lot worse in these pics than in reality.
(https://i.postimg.cc/KRXyWg4k/IMG-20191129-232147.jpg) (https://postimg.cc/KRXyWg4k)
(https://i.postimg.cc/DWzVGrpD/IMG-20191129-232209.jpg) (https://postimg.cc/DWzVGrpD)
The soldering looks okay to me. Where did you get the ICs? I've heard stories of fake PT2399s, and the TL072 seems to be the same *something* as seen in this thread:
https://www.diystompboxes.com/smfforum/index.php?topic=123450.0 (https://www.diystompboxes.com/smfforum/index.php?topic=123450.0)
The TL072 is also backwards in its socket, but maybe the socket is wrong and the chip is flipped intentionally...?
I tried a handful of different PT2399s - some from ebay, and (these) from Tayda.
The TL072 - not sure where it's from, but just now I tried swapping it out with one from Tayda and the sound seemed the same and still unaffected by delay.
I just double-checked the orientation and pretty sure it's is correct.
I had my hopes up for a minute that I might be a simple'ish fix. Those were some good leads worth trying though. Thanks a lot!
edit: by the way, the socket's orientation wasn't deliberate. I just simply wasn't aware until now that it even had a marking on it. Good to know, and I won't make that mistake again.
edit2: also the TL072 I switched to definitely looked different. It had a dot for pin 1, but no half-circle marking on the end.
Looking at the solder side of the board, there are several suspect solder joints. One of the pads leading to pin6 of the top PT2399 looks shorted to the ground pour. Several in many places seem to have incomplete solder. Get a jewellers loupe or something magnifying to give it a thorough inspection.
If reworking the soldering issues don't fix it, it's time to give us some IC pin voltage readings.
Thanks for taking a look. I double-checked (actually triple now) some of the lesser-filled connections and they seemed fine, but I will add more solder next just to be safe and in case I missed anything.
What I think you might be seeing as an unintentional connection to ground is part of the PCB. Not certain I'm looking at the same part as you though. Here are some readings in case anything jumps out as wrong:
These are with respect to the PDF schematic (http://www.valvewizard.co.uk/equinoxII.PDF). I initially did this with a battery but it was draining before my eyes, and I worried that would throw off any calculations or estimations, so this is with a wall wart.
9v = 9.03
Gnd = 0
Q1 (J112)
D = 4.07
S = 4.07
G = 0.01
U1 (TL072)
p1 = 4.07
p2 = 4.07
p3 = 3.88
p4 = 0
p5 = 2.02
p6 = 4.07
p7 = 4.07
p8 = 8.20
U2
p1 = 5.02
p2 = 2.51
p3 = 0
p4 = 0
p5 = 1.22
p6 = 2.48
p7 = 2.03
p8 = 2.11 raises to 2.805
p9 = 2.51
p10 = 2.51
p11 = 2.51
p12 = 2.51
p13 = 2.51
p14 = 2.51
p15 = 2.51
p16 = 2.51
U3
p1 = 5.02
p2 = 2.5
p3 = 0
p4 = 0
p5 = 2.14
p6 = 2.50
p7 = 0.85
p8 = 0.88 raises to 3.13
p9 = 2.51
p10 = 2.51
p11 = 2.51
p12 = 2.47
p13 = 2.50
p14 = 2.51
p15 = 2.51
p16 = 2.51
U4 (78L05)
IN = 8.18
GND = 0
OUT = 5.02
D1
A = 9.05
K = 8.2
D3
A = 0.01
K = 4.07
Can't say for certain without comparing to a known working unit pt2399
But. Can see differences in the pin 7 and 8 in U3 to what U2's are.
Maybe check around there
7 and 8 seem to vary in general, particularly after probing other pins.
Quote from: mark2 on November 30, 2019, 11:59:20 AM
I initially did this with a battery but it was draining before my eyes, and I worried that would throw off any calculations or estimations, so this is with a wall wart.
This strongly suggests that you have a short somewhere. I would bet that something is shorting to ground that should not be.
These ICs are pretty power hungry too though.
But this is definitely a good possibility. I'll look for one again when I get home. I checked all the suspect looking areas already but will be more thorough this time.
If the output mixer op-amp is the second one on the chip, then pin 5 is wrong. You've got 2.02V, and its supposed to be connected to 4.5V.
It's not directly related to the echoes, but it's not right.
Tom
The 2.02V reading is not real. The meter being used to make the measurements is most likely a 1M input impedance meter and is loading down the true voltage. Pin 5 of the OP amp is supplied via a 1M resistor. Vref is supposed to be about 4.5V but is reading 4.07V Vref is supplied via two 100K resistors between 9V and 0V. Ohm's law shows that those reading are correct for a 1M meter.
I suggest that the OP gets an audio probe together and traces where the audio is getting lost or not. But first, lift (desolder ) one end of R4. This will remove the dry audio from going to the output and if there is any delayed audio partially getting through it can now be heard.
I have built one of these reverb units and it is very good.
I had tried using an audio probe but wasn't sure where the problem was because the sound goes through just fine, it's just not affected by reverb.
I'll try removing r4. It sounds like that might make it easier.
I'm going through this with an audio probe.
I got to C4, a 15n mylar cap, and have a clear sound on one side of it, then the other side was a tone. Like as if from a signal generator. Is that normal?
I tried swapping it out, and now I have the input sound plus an oscillating signal on one side, then just the oscillating sound on the other.
I just tried again, and I'm getting a clear signal on both sides of C4 now, it's simply MUCH quieter on the output end.
As time goes on, more and more of that tone or oscillating noise gets introduced.
I unplug, let it sit a few seconds, try again, and now the tone is gone for a while.
OK, this is nowhere near C4, but is an issue. It looks to me as if the holes where you have soldered to pot are grounded connections for soldering the support lugs for a board mounted pot, and are not at all where the electical connections from the pot should go. If you follow through the schematic and the board, it looks as if there are three holes underneath that are unused in your build, but are where the pot should solder to. If you refer to the schematic you will see that all three lugs from the pot should be in use. If you follow the schematic you will see which lug should go the which hole. I have ringed the holes I am talking about in red - hopefully they will show up.
(https://i.postimg.cc/dD8q7x6L/PC.jpg) (https://postimg.cc/dD8q7x6L)
DIY Bass is correct. You have wired the control pot to the mounting holes.
Wire the pot correctly and put R4 back in correctly and this may work properly.
Quotethe holes where you have soldered to pot are grounded connections for soldering the support lugs
Oh my goodness! Wow. Yeah. Thank you.
What makes matters more embarrassing is I keep referring to a big full-page layout of the copper that clearly shows they're both grounded.
I'll fix this tomorrow and report back. Thanks again!
I couldn't wait for tomorrow. I went back and fixed the pot and reattached R4 and....
REVERB!
Thank you all so much for your help.
This problem has had me so focused on the things I don't yet understand (nuances of how this circuit works), and teetering between discouraged and motivated to learn more. All along it's been something that, with a little more attention to detail, was well within my grasp and definitely not over my head.
It pushed me to learn a lot about the PT2399, though, which is good. But lesson learned: Don't immediately jump to assuming the problem lies in the mysteries I don't understand. Chances are good it's an old-fashioned boneheaded mistake.
Works !!! :icon_mrgreen: :icon_mrgreen:
You are quite right. The problem is usually a mistake. Don't feel too bad or silly as everyone makes mistakes. But we do learn from them.
Enjoy the pedal and don't run it from a 9V battery. Two PT2399 chips do use a fair amount of current (about 45mA ). A good 9V battery would last a couple of hours before being excessively drained.
Quote from: DIY Bass on November 30, 2019, 11:42:18 PM
OK, this is nowhere near C4, but is an issue. It looks to me as if the holes where you have soldered to pot are grounded connections for soldering the support lugs for a board mounted pot, and are not at all where the electical connections from the pot should go. If you follow through the schematic and the board, it looks as if there are three holes underneath that are unused in your build, but are where the pot should solder to. If you refer to the schematic you will see that all three lugs from the pot should be in use. If you follow the schematic you will see which lug should go the which hole. I have ringed the holes I am talking about in red - hopefully they will show up.
(https://i.postimg.cc/dD8q7x6L/PC.jpg) (https://postimg.cc/dD8q7x6L)
nice one, well found.
and well done on the "fixing".
Quote from: mark2 on December 01, 2019, 12:35:55 AMDon't immediately jump to assuming the problem lies in the mysteries I don't understand.
This may the THE most important lesson of your career.
My first day at a radio studio, record player didn't play. I had half the studio torn apart. Then elder Stan pointed out a loose plug. Power to the phono preamp. Duh!!
I don't make those mistakes any more... well, any more than 13 times a day. (If I get that far, I take a nap.)
> I don't fully understand yet how the pt2399s are supposed to be adding the delays
_I_ don't understand all the inside and outside parts of the PT-. But at some level we can see it as "Echo Point in a chip". You shout "YO!" and "...oy!oy!oy!" comes back from the cliffs (must be very tiny cliffs inside).
(https://i.postimg.cc/w1MQwBPx/oy-oy-oy-42.gif) (https://postimg.cc/w1MQwBPx)
We want to use this with a guitar. We want the dry signal and the PT- echos mixed to taste. Because the split/mix action and the low impedance of the PT- chip would load-down a guitar, we first buffer the input. Split. One path, Dry, goes to output mixer/buffer. That much worked.
The other split goes to a PT-chip. "Good" is not enough so it goes to a second PT- chip. They are marked short and long. So one is a tile shower, the other is a hard bare cellar. Now we have a long dense "wet" reverb. But because "enough" is still not over-the-top, he adds a "recirculation" path which feeds wet back into the echo path for really-really wet-wet-wet. We *may* not need to delve too deeply in these details.
We need to trim this wet signal from nothing to Lots. The 10k mix pot takes Wet on top and passes 0% to 100% of this to the output mixer/buffer.
When you mis-wired that pot you get 0% to 100% of ground, no-signal. In this case, the problem was nothing to do with the PT- chips but "simple" audio mixing. The mix pot was fed a "nothing". The "...oy!oy!oy!" was right there on the pad NEXT to the pot. And yes, it is natural to look at that 99 times and still not see the slip-up.
Simpler drawing:
(https://i.postimg.cc/Jt0W8N9R/oy-oy-oy-simple-42.gif) (https://postimg.cc/Jt0W8N9R)
Quote from: PRR on December 01, 2019, 03:40:07 PM
Quote from: mark2 on December 01, 2019, 12:35:55 AMDon't immediately jump to assuming the problem lies in the mysteries I don't understand.
This may the THE most important lesson of your career.
My first day at a radio studio, record player didn't play. I had half the studio torn apart. Then elder Stan pointed out a loose plug. Power to the phono preamp. Duh!!
I don't make those mistakes any more... well, any more than 13 times a day. (If I get that far, I take a nap.)
I've heard this expressed as - don't expect a zebra when it might simply be a horse.
Mark2, you owe DIY Bass a (virtual) sausage. Them's the rules.
Great illustrations showing the path, and very helpful to avoid getting lost in the details!
QuoteI've heard this expressed as - don't expect a zebra when it might simply be a horse.
That's good, and very apt!
QuoteMark2, you owe DIY Bass a (virtual) sausage. Them's the rules.
Happy to pay up! That was a good find, and spot on.
(https://i.postimg.cc/w1MQwBPx/oy-oy-oy-42.gif)
there's more oy's there than at the 2000 Olympics.
[edit :] oi!