Hi,
I have build the aquaboy based on Madbeans PCB.
I have used the double delay time board with two V3205's, MN3102 and NE570.
It is working, but I have a delayed signal that is louder then I think it should be.
Mix at noon puts the delay signal already way above the dry signal. Also I have some distortion on the delays.
Also the feedback starts oscillating at 1 o'clock. Way too early.
I checked all components and readings. R31 is 22k in my build.
I accidently turned Q4 around and although that should not work it brought the signal to the right level and also the feedback was reacting better.
That made me think that the delay signal is too loud because either the signal from the NE570 to the V3205 is too loud (which would explain the distortion) or the signal coming from the V3205 to the NE570 back is too loud (which would not explain the distortion). I do know that the delay should not be completely clean but this distortion is too much. Could it be that we are driving the V3205 (in my case even two) too loud in this build??
All voltages read as posted elswhere. There is only one deviation in reading: NE570 pin 10+ 11 read 3,5 volts where it should be 4,1 volts.
Thanks for the help.
Best, Marc
In the mean time I have been working on this build again. Checked everyting three times.
Then adjusted all the trimpost using the method that is explained on BYOC video's.
The delay works well and I can adjust all trimpost so that I have the least noise and the best signal and delay time.
But the delay signal is still too loud. I think I will have to tame the signal from Q2 going into the V3205.
Any suggestions or other remarks?
Thanks, Marc
Marc: to reduce the overall delay level try this: socket R33 and try increasing values there, say 68k, 100k, etc. Or, while it's socketed, use a 250k pot set up as a variable resistor (lug1 on one end, lug 2 on the other) and fiddle with it until you get the right mix of delay and clean signal. Then measure the resistance across those two lugs and use the closest value fixed resistor you have.
Hi Brian,
I will try your suggestion but if I am correct then if I follow your suiggestion I am changing only the balans dry/delay.
I was wondering if I could change the delay level in another stage.
I would prefer to change the level going into the V3205's. Somewhere around Q2.
In that case I can reduce the distortion in the delay and at the same time reduce the feedback that starts ocillating before "noon".
Thanks for helping.
Marc
Gotcha. From the top of the circuit, you have a gain of 2 @ IC1A going into the compander, then some filtering and another buffer at Q2 to bring it back to unity. To reduce the signal at his point, you could leave Q2 out, and jumper the base to emitter. Leave the collector unconnected. You might jumper out C15, too, but I'm not sure.
This should knock the signal down so it's not so hot going into your v3205. Mind you, I have never tried this, so I cannot guarantee it is the best solution...but it might be worth your time.
Alternatively, you could try either a variable resistor or voltage divider right after C15 to control the level into the BBD.
Hi Brian,
That was an easy one to try as I have socketed my transitors. I jumpered Base to emitter.
It does bring the signal down but it is too low. I also tried jumpering C15 which had no effect.
I experimented by doing the same on one of the other transitor stages after the v3205 but without the right effect. Funny thing is that if i turn either Q2,3 or 4 around then the signal is of better quality, the level is right and the feedback is the right amount???
Any other suggestions? Can I change the amount of gain in the Q2 buffer??
Thanks, Marc
Hi,
I missed your post suggesting the variable resistor or voltage divider. I still have to try that.
In the meantime I worked with an audio probe.
If I compare the dry and the delay level at R36 and R33 then it is very clear that the whole delay line is way louder then the dry line. (but we knew that already;))
Could it be that the NE570 is amplifying the signal too much? I have swapped the NE570 with a NE 571 I have and they give the same result so I guess the NE570 itself is ok. As I wrote in my previous post: NE570 pin 10+ 11 read 3,5 volts where it should be 4,1 volts. Is that indicating something??
Marc
Hi Brian,
I followed up on your siggestion ib ptting a variable resistor after C15. I used a 100k pot and found that with 56K I had an effect level that was cleaner and better in balans. Now the effect is at unity when the mix knob is at noon with a beautifull effect sound with no distortion.
The feedback however is still too sensitive. It starts feeding back at 2 -3 o'clock. I did not want to bring down the level at C15 more to tame the feedback as it would make the delay level too low.
The only thing I have to do now is to bring the feedback down a little so it will start feeding back at around 3-4 o'clock. I remember somebody writting about that so I will dive into that.
Thanks, Marc
Great! This is a good tip, then for balancing the delays.
For feedback control, socket that 22k off the feedback pot. Try increasing the values incrementally. That should do the trick.
Hi Brian,
I already figured that one out and used a pot to see what value to use and ended with a value of 30k.
So, to conclude I now have a working Aquaboy with double delay time and modulation board.
For future reference:
I have used V3205's, NE570 and MN3012. I added a resistor of 56k from C15 to the junction of R15 and the BBD to bring down the level going into the delay (and to control the distortion). I used a 30k for R31.
The delay is quiet and and clock noise is not audible.
Thanks for putting me in the right direction and for the pcb.
Best, Marc
That's great Marc.
Next week I'm going to create a 'mods' section on the forum to collect stuff like this into one area. This is very helpful info for future builders of this project.
I boxed the delay up today. I have build a few before like the PT 80, the Femtoverb and the Echobase. This could very well be the best of the them now although I have a long very good relationship with the PT80 that I use a lot on stage.
Highly recommended.
Thanks for helping.
Best, Marc
I do have some trouble getting the mod board to work right.
The modulation board works. I can get deep modulation and a high speed if I want but at certain settings (towards the extreme) the delay suddenly cuts out. Then I have to move either the delay time knob or the modulation board to get it working again. Even if I bring the clock time down (which was suggested before to be able to use the extremes of the modulation) I have this same thing happening. I already added extra resistance (a 50k trimpot) after the modulation depth pot to control the modulation depth but that does not solve this cutting out.
It does only happen in the combination with the mod board. Without, the delay works fine.
Also the speed extreme is to high to be usable. Where should I start to solve this? As I see I have three problems: the Mod depth is maybe too much, the mod speed is too high and the delay cuts out on certain modulation depth settings.
Thanks for helping.
Marc
Hi there, could I use a larger value resistor in place of R15 instead of adding another to limit the volume to the chip? The last one I had built was distorting too much, no matter how I calibrated it (double delay mod).]
Cheers
Andy
Quote from: hoyager on September 13, 2010, 10:57:19 PM
Hi there, could I use a larger value resistor in place of R15 instead of adding another to limit the volume to the chip? The last one I had built was distorting too much, no matter how I calibrated it (double delay mod).]
Cheers
Andy
Hi Andy,
My experience was that the BBD was getting too much signal and that was causing the BBD to distort.
I replaced R15 with a trimpot to find out at what value the signal was loud enough and clean enough. Then I removed the trimpot and measured the value at wich the trimpot was set and used that value for the reistor at R15. So yes you can substitute R15 with a different value, but you have to find out what works for you or follow my suggestion in a few posts back.
Success,
Marc
Cool, thanks. So you effectively have a 156k resistor for R15?
The last one I built with mod and double delay had a similar probllem where the delay seemed to cut out at the fastest setting but it ended up being a kindof distortion which was quite cool. So you're not alone there!
The higher settings were kind of bunched toward that end of the pot, was that the case with yours too?
Andy
Quote from: hoyager on September 14, 2010, 11:24:44 AM
Cool, thanks. So you effectively have a 156k resistor for R15?
No, I think I ended up with 56k (I used a 100k trimpot to find the value and measured 56k when I removed it and so I used that value (56K) for R15) But it might be different for you.
Quote from: hoyager on September 14, 2010, 11:24:44 AM
The last one I built with mod and double delay had a similar probllem where the delay seemed to cut out at the fastest setting but it ended up being a kindof distortion which was quite cool. So you're not alone there!
The higher settings were kind of bunched toward that end of the pot, was that the case with yours too?
Andy
No I noticed no distortion at the cut out point of the mod depth. I have not found the best solution yet. I now have a trimpot connected to the Mod depth pot to limit the range up to the point that it does not cut out. I still have to find a better solution. Untill now no one responded to that question regarding this problem yet.
Yes the higher settings were bunched towards the end of the pot with me too. I guess replacing it with a log pot would help....
Marc
Awesome thanks, I think i'm gonna roll with a 68k, seeing as I have those, and with the stock 100k it wasn't unusable.
So, B is Linear, and A Log? I've read there's been some swapping of codes... The build pdf sepcifies 1MB, that should be log?
1MB is linear. "A" is audio (log).
Cheers, so is it supposed to be like that with the short times bunched up? I mean is that a design feature?
Out of interest, I've built a second pedal complete with dd and mod boards, but used all green / brown mylar caps, and the blue HV ceramic caps and it has come out much darker than the one before with Wima film caps and regualr ceramics. The feedback tick when the knob is full from nothing is much much deeper as well. I quite like it... although I used the multi layer cermic caps at the 330p values and I think they are causing a bit too much overdrive / distortion, its either them or the blue HV ones, I'm going to swap them back gradually, to see thats the case
Is it possible to have a switch cut the dry signal out, or the mix knob cut it out, so the pedal can be used as a send effect for dubs?
Andy
If you use a switch in place of R36 you can kill the dry signal. You just need to lift out the signal coming from IC1A.
Not sure what you mean about the shorter delay times bunching up. With a 1MB the time transitions are pretty smooth in mine.
If I read the last few posts I think we are mixing two things up:
I my post I said that the higher settings were bunceh up towards the end of the pot meaning the higher modulation rate settings being bunched up on the speed pot. I was not talking about the delay. I already tried a log pot wired in reverse and that spreads the speed evenly. My recommendation would be to use a reverse log pot. I am still fiddeling with the mod board so maybe this decission will change later.....
In my experience the delaytime pot on the aquaboy is quit evenly spread over the range of the pot.
(I have also a double delaytime board and V3205's)
Marc
Oh! My bad. I have two each of the mod and double delay time boards form Marcus. I really need to get those built and added to my existing Aquaboys so I can speak intelligently about this :'(
Hi, I've just built another version and am having what sounds like the same problem with the signal going to the chips too hot.
I've tried replacing R15 with a smaller and larger value which has no effect for me. Swapped out all transistors and all chips, checked all resistor values and caps. Have adjusted all the trimpots for optimimum too. I've got 1.5k at R19 and I'm using green caps for all the large values 47n up and a mix of box caps and mylar for the smaller, tone shaping values
The only thing I haven't tried is jumpering out C15 and am wondering if its possible do that leaving it in the circuit and bridgeing the connections?
I feel like a cap is faulty or one of the diodes, somehow, and there is too much current flowing and/or the chips are being overdriven. I think it would sound ok if played aggresively, on a guitar, but its not as nice or subtle as the other versions I've made and as soon you play chords into it it becomes quite obviously overdriven.
Are there any other components before the delay chips I can check?
Andy
Here are some voltages, maybe be a better picture
v3205 x 2 DDT mod board (pretty much the same for both)
1. 0 (2nd chip)
2. 3.35
3. 3.95 (4.24)
4. 4 (4.25)
5. 7.4
6. 3.38
7. 3.7
8. 6.35
This from a previous version I've built which is cleaner (very similar in cap choice (green mylar for large values and an assortment for smaller values)
1. 0 (2nd chip)
2. 3.38
3. 4.08 (4.54)
4. 4.11 (4.52)
5. 7.35
6. 3.36
7. 3.65 (4)
8. 6.3
The only thing funky on both of them is a 120pf instead of 100pf at C21 after the clock. The cleaner version also has a 47k at R3 instead of 475k, doesn't seem to make any difference somehow... Also on this version I had the diodes swapped, one for the other, and it worked and sounded pretty good (they are switch to spec now though)
Here are some audio comparisons, cleaner older version then newer darker version
http://littlemp3.com/daf48c7
http://littlemp3.com/b09750c
http://littlemp3.com/e8b8f72
http://littlemp3.com/d2e661e
http://littlemp3.com/76d4eb8
So obviously the new one is voiced darker, but I was trying to disguise the overdriven sound. The darker one has alot more bottom end content even when playing higher notes
Am I just imagining that is sounds bad?
Andy
Hi Andy,
The clips sound like they should mostly.
I was thinking of a few things to check maybe, if you have not done that yet.
- the signal going in your delay could be too loud, using that piano.
- I read that you adjusted the trimpots as they should be, but how did you do that? Did you use a scope or did you do it by ear. I could imagine that doing it by ear with, for example a rhodes sound, is not going to give the right comparison. Adjusting the trimpots is a pretty precise job. That is why some use multi-turn trimpots in this build.
- if replacing R15 with either a lower or a higher value is not changing anything then that makes me think that your signal driving the delay is too maybe way too loud, as it should really chance with a different value. (Or you are chosing values that are too close to one another of course).
I hope this helps.
Marc
Hey thanks for your suggestions. The rhodes is perfect for revealing any distortion because the signal is almost a sine wav and its def not going into the delay too hot. As far as trimpots go, I'm absolutely sure they're dialed in, I mean my ear is a lot more sensitive than my hand holding a screw driver.
I think I was after a technical solution. I've tried replacing 2 of the 220n green mylar caps with wima box and no difference there. I've added a 2.2k resistor to the 1.5k at R19 to giving about 900ohms which *seemed* to clear it up, but only slightly
I've tried a 10k at R15, should be 100k, but it didn't affect the distortion, just the amount of feedback. It seems slightly quieter (just listening to the wet signal with feedback dialed down) but the distortion is still at the same level.
I'm also thinking maybe I should put in 4 new mpsa18 incase any of them were damaged during my previous swapping, or they are faulty? I think someone else experienced this too?
here's another example with the same sample at the same level going through the *questionable* delay, then the one that works well, all with the same settings; wet only signal, level on full, modulation off,
http://littlemp3.com/c632a29
Andy