ABDX with double delay board

[gitaar0]

I started a new thread on this although some of it came up in my questions on the ABDX moulation.

With a double delay board it works with 3 mn3005s.
I have used the double delay board for the original aquayboy wih the values for that.
The delay is quite a bit louder with that board. Brian already suggested changing the trimpot or R39 to deal with the louder repeats and the resulting early oscillation.

What i also notice is that with the delay board the repeats start to slowly sound worse almost as if there is a gate on the repeats that cuts of part of the sound.
This is not the case without the double delayboard and just 2 MN3005's. In that case the delay repeats an slowly gets darker.
The voltage on my MN3005's is normally 14.45 volt (14.80 at the roadrage board).
With the three MN3005's the voltage goes down to 14.30 volts.
Is the voltage dop is what causes that gating? R52 limits the voltage, should I change this?
Or is something else causing the gating...?

I changed T1 trimpot to 100k and changing R39 to 130k to deal with the louder signal and the early osillation on the feedbak pot but that makes the repeat signal gating worse. Is this because I am now adding resistance after the feedback pot and also after c31? or just beaues it is a lower signal?

The ABDX sound great with just two mn3005 but it would be great to get it working with the double delay board.

Thanks,
Marc

[madbean]

Are you using the double delay board from the Aquaboy 2012 docs? I just looked at these again and noticed there is a mistake on the board. R1 and R2 are 100k pulldown resistors for each output of the BBD. But, it looks like I have both attached to pin3, rather than one to pin3 and one to pin4.

If you are using that double delay board, move either R1 or R2 so it connects with pin4. I will also correct the mistake ASAP on the doc.

I'm not sure that is what is causing the gating. But, I also have not tried the double delay board on the ABDX. The clock is meant to drive two BBDs, and it may just not be suited for three or four without adding a second clock or some kind of clock buffer. I just can't speak too intelligently about it not having tried this myself.

[gitaar0]

the double delay board dates back to 2010. It is the one that was made by abfackeln on FSB.http://www.freestompboxes.org/viewtopic.php?f=13&t=8376&hilit=double+delay+board&start=100


I understand that it might be that the clock is not suited for three ic's. How do I check that?


Edit:
I tried to bias some more and found that that has an influence on the gating as well. So it needs very precise biasing, even more then with three mn3005s. I found a position where the total sound works well but I have more distortion with the three then with the two ic's. Maybe I should limit the signal going in the third mn3005. How do I do that?

I limited the signal going into the other two mn3005's. I now have 39k for R55 and 100k for R27. After that I rebiased and that cleans up the signal but I think it could still be cleaner. Setting r13/14 to 10k cleans it up enough.(even 20k did leave some distortion) It seems that that has the most influence. I noticed that with this value I could bring down r55 to zero with no extra distortion.

I bypassed R52 so the voltage on all three ic's is 14,80volts. That seems to have an influence on both the gating and the distortion as well. Can I just jumper R52?

[gitaar0]

after tweaking some more my conclusion for now is that running three MN3005's from one mn3101 is not working distortion free.
If there are suggestions I am open to that?

With two MN3005's I have a very good signal but with r13/14 at 10 k. any value higher introduces distortion. Maybe it happens sooner in my set up as I work with humbuckers (altough these are low output compared to others).

I still have too much oscillation on the feedback pot. Trimmer T1 is 100k. If I raise R39 I get a more distorted signal. Somehow I have the impression that introducing more resistance at the r39 deteriorates the signal. Should I change the feedback pot or put more resistance right behind that pot?

[Scruffie]

The filtering and circuit was never designed for that much delay, so it's not too much of a surprise really. BOSS would have designed it to give 300mS reliable delay, with 2 x that much, it probably manages fine, 3 x though?

Also you will need more precise biasing with more BBDs, as if one is slightly off, the next one in line will be working with a slightly off signal and then degrading it more.

The clock (MN3101) can most likely cope with it fine (not really designed for it but...) you'd be getting a lot worse than feedback and distortion if it wasn't, more like skipping and failing to delay as the square wave became rounded. If you want to check it, a 4049 buffer is pretty easily wired up on a piece of vero or perf and injected in to the circuit. Look at the current lover project and it should show you what you need to hook up, 2 points in to the 4049, 2 points out.

Humbuckers wont help, BBDs aren't the best devices for higher output signals, lower output or not.

Post the schematic so I can see which resistors you're fiddling with.

[gitaar0]

Hi scruffie, thanks for replying.

The schematic is here http://www.madbeanpedals.com/projects/Aquaboy/docs/AquaboyDX.pdf

I know I am stretching what was intended with the design. It just sounds very good wih two bbd's so why not add one more

I hear what you write about the clock, i will check out the buffer idea.

[Scruffie]

R13 & 14 should be about 16k at 15V to bias the 570 at half supply I think... but without checking you may wish to measure the difference in voltages and aim for half supply with your values.

Introducing more resistance at R39 will weaken the repeats as the signal struggles to go back through the circuit so will deteriorate the signal/ lead to repeat gating. You can adjust C31 to let less signal through to avoid excess feedback.

How are you getting your 15v? Not sure what the current draw of this circuit is but it may explain the voltage drop from the 3rd MN3005 and could lead to the signal cutting in and out with the modulation, the LED probably pulling a fair amount of current, taking it out of circuit may help. That small a voltage drop shouldn't cause too many issues with an MN3005 (you can check the graphs on the datasheet for volts vs ####).

As mentioned, precise biasing is going to be critical here. How have you set up your extra MN3005? Does it have series resistance at the input as well like R27 & 55?

[gitaar0]

Hi scruffie,

I will check r13/14 to see what voltage i am wih the 10k i am using now ( and what is in the original)

Your remark on r39 is what I noiced while playing with that value. That is why I changed r39 back to 5k1. I will change c31 although that might have to wait till I have the signal flow sounding good.

I am powering from a roadrage, lt 1054 based voltage converter. I have used that before and have sometimes struggled with the voltage dropping when using this set up. Maybe that is the case again, with the third bbd being just to much. But as the bbd works also on 9 volt it should not be distorting, i would think.
I will see if i can set up a different power supply.

Thanks

I used the double delay board that was originally designed for the first aquaboys at FSB. See earlier post for a link.
It is bassically the set up of the original bbd with the cancel and trimpots and the three 100k resistors. No inout limiting AFAIK. How much would you limit it?

[Scruffie]

Hi scruffie,

I will check r13/14 to see what voltage i am wih the 10k i am using now ( and what is in the original)

Your remark on r39 is what I noiced while playing with that value. That is why I changed r39 back to 5k1. I will change c31 although that might have to wait till I have the signal flow sounding good.

I am powering from a roadrage, lt 1054 based voltage converter. I have used that before and have sometimes struggled with the voltage dropping when using this set up. Maybe that is the case again, with the third bbd being just to much. But as the bbd works also on 9 volt it should not be distorting, i would think.
I will see if i can set up a different power supply.

Thanks

I used the double delay board that was originally designed for the first aquaboys at FSB. See earlier post for a link.
It is bassically the set up of the original bbd with the cancel and trimpots and the three 100k resistors. No inout limiting AFAIK. How much would you limit it?

Yeah C31 can be left until you've solved the main issues.

Yes but i'm saying the modulation is pulling more current as it sweeps and could be pulling on what's available to the BBDs and gating out. It's just an idea anyway, probably not what's happening, i've never looked in to the current draw of a single BBD but a single LT1054 may be at its limits with the circuit.

I'd just start with 1k for limiting, it'd be worth using a 25k pot though probably just to make testing easier as that's purely a speculative guess.

[gitaar0]

this is what i measure at my ne 570 with 10k at r13/14 pin 7 reads 3.03. With 33k at r13/r14 pin 7 reads 5.80 and at 47k pin7 reads 7.55

Pin 13 reads 14.80 so half supply would be 7.40 volts.
But I do get more distortion then so should it really be half supply voltage?

I also took out r23/24 and put an extra cancel trimpot there so I could  dial the right 'cancel' value with a scope. That made quite a difference in sound, better for sure. Maybe this should be part of the design anyway.

I took out the lfo ic to see if that made a difference in the voltage drop. It did not.

What I did notice is that when I connect my 9 volt power supply (a one spot) to the roadrage that is powering the ABDX the voltage takes a minute to get fom around 12 volt up to the 14.80 volts. Maybe the roadrage is not really happy in ow much it has to deliver or is this normal behaviour?
I have had some trouble wih another roadrage before ( that was or a dirtbag delay) and I am not so sure if it really delivers enough power. I should set up a 15 volt supply without a voltage doubler to check this.

[Scruffie]

this is what i measure at my ne 570 with 10k at r13/14 pin 7 reads 3.03. With 33k at r13/r14 pin 7 reads 5.80 and at 47k pin7 reads 7.55

Pin 13 reads 14.80 so half supply would be 7.40 volts.
But I do get more distortion then so should it really be half supply voltage?

I also took out r23/24 and put an extra cancel trimpot there so I could  dial the right 'cancel' value with a scope. That made quite a difference in sound, better for sure. Maybe this should be part of the design anyway.

I took out the lfo ic to see if that made a difference in the voltage drop. It did not.

What I did notice is that when I connect my 9 volt power supply (a one spot) to the roadrage that is powering the ABDX the voltage takes a minute to get fom around 12 volt up to the 14.80 volts. Maybe the roadrage is not really happy in ow much it has to deliver or is this normal behaviour?
I have had some trouble wih another roadrage before ( that was or a dirtbag delay) and I am not so sure if it really delivers enough power. I should set up a 15 volt supply without a voltage doubler to check this.

Look up the 'Compander Cookbook' half supply on the outputs should provide the most headroom according to its application notes and it makes sense, although perhaps more value tweaking is needed to get the output of it right at that point.

But... 10 & 11 should also be half supply so try adjusting R37 too.

Yeah, try it with steady 15V, always good to rule out anything. I've never measured charge pumps on start up, does seem a bit odd that it would need to charge for any longer than a couple of seconds though.

Makes sense on the cancel trimmer, gives the next BBD better signal to work with, however this all seems a lot more finnicy than it should be really.

[gitaar0]

Hi,

I will try adjusting  r37.

The extra cancel trimpot seems much but it did make a difference. Guess I am at a very precise tour now.
I love a good sound

I will only be able to get back at this by the end of the month.
So it will have to wait.

Thanks, marc

[gitaar0]

Took a while to get back to this.

I first worked on adjusting the in and out voltage  of the campander to half supply by adjusting r13/14 and r37 as Scruffie suggested. Then i changed r55 to 50k and r27 to 100k. (i am working with humbucking pickups remember; this may not be needed with singlecoils) Then i readjusted the value of t1 and r39 to he original values. That with quit some fiddling with the different biases cleaned it up a great deal. I find that it is easier to bias at the wiper of the cancel trimpot by the way, when using a scope. You can adjust the cancel pot and the bias pot from one measure point.

I noticed that any resistance extra in the feedback loop greatly reduces the quality of the repeats.
So i am trying to avoid that. It means that the feedback is oscillating now at about 1 o'clock. I will have to find a solution for that other then raising he resitance of t1 and r39. I will experiment with the suggestion of changing c31.

Now to the double delay board.
I did put a 47k resistor at pin 7 before the last bbd. Now after biasing and biasing (scope and ears) it works very well. It is quite a bit darker then with two bbd's. I will have to see if that is worth the extra delay time, or i will have to adjust the filtering

Because of the longer delay, finding the right shorter delays is more precise. Maybe i should change the delay pot to log?
Same with the feedback pot to put the oscillation later in the pots sweep?

[Scruffie]

Took a while to get back to this.

I first worked on adjusting the in and out voltage  of the campander to half supply by adjusting r13/14 and r37 as Scruffie suggested. Then i changed r55 to 50k and r27 to 100k. (i am working with humbucking pickups remember; this may not be needed with singlecoils) Then i readjusted the value of t1 and r39 to he original values. That with quit some fiddling with the different biases cleaned it up a great deal. I find that it is easier to bias at the wiper of the cancel trimpot by the way, when using a scope. You can adjust the cancel pot and the bias pot from one measure point.

I noticed that any resistance extra in the feedback loop greatly reduces the quality of the repeats.
So i am trying to avoid that. It means that the feedback is oscillating now at about 1 o'clock. I will have to find a solution for that other then raising he resitance of t1 and r39. I will experiment with the suggestion of changing c31.

Now to the double delay board.
I did put a 47k resistor at pin 7 before the last bbd. Now after biasing and biasing (scope and ears) it works very well. It is quite a bit darker then with two bbd's. I will have to see if that is worth the extra delay time, or i will have to adjust the filtering

Because of the longer delay, finding the right shorter delays is more precise. Maybe i should change the delay pot to log?
Same with the feedback pot to put the oscillation later in the pots sweep?

Glad you got it sorted!

Double delay bypass toggle?