Aqua Boy with v3205 's

[gtr2]

No progress made...

I did get the voltage divider to work lowering the signal into the bbd.  No change in the distortion.  So I put it back together stock eliminating the 4049 buffer and I didn't notice that it was any worse at distorting.  So then I pulled Q2, socketed it, and subbed in some lower gain npn's.  Even with the reduced signal into the bbd's they were still distorting.  Each time a change was made in all the above a new calibration was done by the way and only one change was done at a time.  I'm wondering if switching out to multiple turn trim pots would help me dial this in more. (at least for the bias).  Maybe the bias range with two of these is so narrow that it's virtually impossible to find the sweet spot with little dirt.   

I did get an oscilloscope for FREE, but it's missing the trace.  So I'll probably working on that instead for a while so I can get a better look on what's going in with this build.  It's an old tektronix 7904 about 30 years old.  I guess they were $28k back in the day.  Ouch!  It's a beast.  Big and full of mojo...

Josh

[gitaar0]

Josh,

Check the Byoc video on how to set the bias and cancel of the bbd's. It is very precise and should be done with a scope or audio probe on the output pin of the bbd's. For each separately. If i read your posts then I keep having the idea that that might not be done right.

I read that you read my post on r15. If changing r15 deos not do anything then there is something wrong as it has a huge influence on the signal going in the bbd's.
I have build this with two v3205's and it is working really well with quite a long delay time.

Best, Marc

[gtr2]

Thanks Marc,

That's how I've been calibrating from the start of this (actually before this project).  I use an audio probe.  I'm pretty good on dialing the old analog stuff like my old tape delays and other analog delays.  These v3205's are just so much more finicky with everything.

I've been fooling around with some stuff like the voltage to the bbd's etc along with adding in a resistor right after c15 which forms a voltage divider with r15 to the bbd input.  I also added in the 4049 buffer again.

I am really, really, close now with this arrangement.

I'll keep everyone posted.

Josh

[gitaar0]

Well in that case my suggestions were maybe not necessary. It just is strange that mine turned out really well ( and i am very speciic with my sound stuff. I use all my pedals in my work as a recording and touring guitarist) and yours just keeps not working well enough. Hope you get it sorted out. i will keep following this.

Best, marc

[madbean]

I'm building up a dual delay too to see if I get the same result. Hopefully I will be able to add something useful to the discussion.

[gtr2]

I'm building up a dual delay too to see if I get the same result. Hopefully I will be able to add something useful to the discussion.

Good, because I'm really pretty frustrated at this point.  I just packed my rats nest up for now before I did something stupid.  There's no point in working on it with my current mindset.  Um...just ask one of my old computers 

I upped my voltage to the bbd's to 8v and cut the input signal with an additional 100k resistor ( I think, I tried a bunch) right after c15 forming a voltage divider with r15.  This worked but it also cut the delay signal to much.  I had to put a jumper in place of the resistor at r31 so I could get more than 2 or 3 repeats, but the repeats are just to far back in the mix.  The delay loop needs boosted somewhere to compensate.

In my findings anyways..I couldn't calibrate the 2nd bbd to get an output signal without using the 4049 clock driver buffer when I decreased the input signal.  I had to use the buffer to get any delay.

O well, I feel pretty defeated at this point.  I'm just glad this one was for me and not a friend.

Josh

[madbean]

I haven't wired up the DD board yet, but I at least verified my new build with a single v3205.

Couple more notes on the single v3205 version:
A 5v regulator is not a good choice for feeding the BBD. The headroom is just too low. I was able to clean it up some by using a 470k on R15, but not much.

Using a 1k5 resistor for R19, as suggested in the build doc, really works well. Hitting the BBD at about 7.5 - 7.7v gives it plenty of headroom. I was able to dial in the clock and bias easily by ear to obtain acceptable delay times with no clock noise. I did leave in the 470k on R15.

At 7.7v, the 22k feedback resistor is a little low. Oscillation occurred at just over halfway on the Feedback pot.I haven't subbed that yet, but something higher like 33k or 47k should do it. At 5v on the BBD, the 22k did work better.

Lastly, using the modulation board seems to bring out a lot more clock noise at high speed and depth settings. I was able to dial this out by setting the clock trimmer back a bit.


Anyway, I'll give my report on the dual thing soon.

[madbean]

Okey doke, my experimentation is done and the results are pretty good.

I got the best results by doing the following (BTW all this was done by ear---no scope):

Run the circuit at 12v regulated. While the Clock is rated for up to 12v max, the Compander indicated 8v max, according to the datasheet. However, I had no issues running the board at 12v, so I suspect there is some give there.

The reason for running it at 12v is the expanded headroom helps a lot with the dual delay set-up. In fact, it looks like running the clock at a higher voltage allows you to cook the BBD a little to get that extra headroom needed to offset the noisey dual delay setup. I found that I could easily run the BBD at 9v with very acceptable results.

First off, it's going to be easiest if you first test your build as a single delay and bias it up as best you can. Having the Clock and Bias trimmers set on the main board first for a 300ms delay makes adjusting the Bias on the daughter board way easier. Once I plugged my dual delay board into the socket, it required minimal adjustment to the second bias trimmer to get the optimal delay. Adjusting the second cancel trimmer allowed me to dial out a little noise, as well.

So, for the power supply, I used a Road Rage board, opting for a 12v regulated output. This plugs into the 9v of the main board. Then, for R19, I used a 9v regulator. This actually yielded about 8.8v in my case, but that was fine. I left R15 at the 470k I subbed in earlier, and upped R31 from 22k to 150k to control the self-oscillation.

Following this I was able to around 500ms or so of very nice delay. I say 500ms, but I don't know how exact that is....it's a guess.

At maximum delay, it's obviously going to get kinda dirty. However, it's very dynamic sensitive. So, playing softly, I got very distortion free results. Picking/strumming hard essentially 'overloads' the BBD some, but even then it was not un-musical. The distortion is more like the DMM overload, and not hard/nasty clipping.

Hopefully this gives you some new methods to try, Josh. If you've soldered the dual delay board to the main board already, you'll have a bit of a time desoldering all that. For mine, I have a socket on the main board. On the dual board, I took resistor leads, folded them in half and soldered four of those to the 4 pairs of pads on the DD board. Then I snipped the connections on the top side and trimmed the leads on the bottom to the same length. So, it's essentially "plug and play" this way...you can always remove it and put a single BBD on there at a later time.

[gtr2]

Thanks a lot Brian 

Yes, I soldered the DD board with old resistor leads and didn't think to socket the spot on the board to put the leads into.  That's a great idea, because If you have to get to anything under the DD board you can pull it out as well.

Did you happen to do a comparison with the double delay before and after your modifications?

I'm gonna have to get some regulators, I don't know if they carry them at "the shack".

I also noticed that the byoc analog delay project uses the mps6521 in what I consider Q2, Q3, and Q4 with a mpsa18 in Q1.  I may try this too depending on my results.  The circuit is pretty similar to the aquaboy.  It doesn't help that I only have humbucker guitars either.

This definitely gives me a good direction to go and I really appreciate the time you've put into this Brian.  You've really made this a great place for DIY!  Plus the people around here are cooler 

Depending on if I have the parts or not may vary the time on my report back, but I'll let everyone know when I do.  All things considered, I've learned a lot about bbd delays.  I collected as many bbd schematics as I could find and studied them for "clues".

Josh

[madbean]

Josh, if you don't have a 9v regulator, use a 2k2 resistor to get about the same result (if running at 12v).

Here's a pic of my set-up.



[attachment deleted by admin]

[gtr2]

Ok, thanks.  I'll fool with it tonight again.  I can power it with my pedal power plus at 12v for now.  (I forgot I could do this) If this works out I will build up the road rage board for it.  (I got one as a present  )Then it will be much more flexible for switching between boards with different power supplies. 

I'll sub in the 2k2 or whatever gets me at 9v for the bbd.

Josh

[gtr2]

Ok, powered the pedal up at 12v.  But I ramped it up it slowly using my test rig voltage pot so I knew I wouldn't overpower the bbd's.  With the 2k2 it put me right at 8v.  I still found unfavorable distortion, but it was improved!  So from there I added a 37k resistor after c15 with a 100k at r15.  I wanted to tweak it a little but it finally had the correct sound for an analog delay.

So then I went to eat.  When I came back and powered everything up I had no more delay.  I'm unable to get delay at even the stock settings.  I did not slowly ramp up the power this time and my power source put out about 12.6v.  Did a surge fry the clock?  Is there any way to check this besides trying a new one...  I didn't smell anything funky.  The data sheet also says this "Incorporates a diode to protect the IGBT gate at power on"  If it needs replaced would a mn3101 work.  It's rated for -8 to -16v.

Josh

[bigmufffuzzwizz]

Is there any way to check this besides trying a new one...  I didn't smell anything funky.

I believe with a logic probe you would be able to figure out if its burnt.

[madbean]

Try replacing the compander first, if you have an extra. You should be able to use a NE570, if you have one.

I tried mine at 15v briefly, but got no delay. However, my chips were undamaged and worked again after running off 12v regulated.

[gtr2]

Hmm.. I don't have an extra but I've got a small bear order that hasn't shipped yet so I can probably add some extra stuff on...

The compander is rated up to 18V so I didn't really consider that.  I'm gonna get that and a clock anyways.  I've got my dd board separated now since I have to socket where it goes in.  There's nothing more fun than unsoldering, 

Josh