Cavedweller dirt in repeats...

[greysun]

hey there!

I finally just built a cave dweller. Full disclosure: I etched a custom version because I have a bunch of .1uf poly caps that are 7 through holes wide (aka, HUGE) and needed to use them up. I also wanted something to test out the press-n-peel method of etching before I make a custom switch. This will go into a 1590B. (an aside: I hate etching boards, so it's rare, but this went well. The vast majority of my pedal board is madbean fabricated, though, hehe).

So it's built - it functions - it's pretty cool, actually, BUT... The initial swipe of the guitar is pretty clean, but the repeats seem to have a little dirt or something in them, especially when the delay is longer. I got Q1 drain to 6.24V with a maxed out trimmer. I'm using a green diffused LED with the PT2399. I swapped the PT2399 already just to be sure. Pins 3 and 4 are both connected to ground per the schematic (initially I only had pin 3 connected to the LED, but then connected it also to pin 4).

I read that it has "dark tones" in some reviews, and I know it's supposed to be a more lo-fi delay per the schematic. It's pretty neat either way, and this is a gift for my voice instructor to use while teaching, so it will work fine in that regard, but am curious if there's something I can do to clean it up a bit.

If anyone has any thoughts, or if this should be in another area of the forum, let me know. And, as always, thanks in advance for any help!

[alanp]

If you want clean repeats from a PT2399 in a relatively small board, then your best bet is the Deep Blue Delay clone board. The Cavedweller really isn't built for clean, at all.

Brian's version, _Sea Urchin_, is here -- http://www.madbeanpedals.com/forum/index.php?topic=2832.0

[midwayfair]

The PT2399 distorts above delay settings of 200ms. You can get it relatively clean up to about 350mS with aggressive bandpass filtering. Increasing R9 would decrease the treble in the repeats. Decreasing C2 to, say, 10nF would cut bass in the repeats.

When you start looking at longer delay settings like 500mS, you are either filtering extremely aggressively to get clean repeats or you're living with the fact that the 2399 is a dirty chip. Merlin Blencowe wrote a good piece about it years ago but I can't find it now.

Despite what Alan says, the current version of the CD has about the same delay path stuff as the DBD or most others, it just has a greatly simplified dry path. The older CD was pretty much "live with it."

[greysun]

This is about what I thought, and I think I'll leave it as is. Don't get me wrong - this is a cool pedal! I was playing a version of a fingerpicked Owen song (playing possum for a peek, if you're curious) on a jazzmaster through it, and the effect was pretty amazing.

On my own board, I use a 1776 multiplex and love it. I think I only use one of the settings, so I may soon get the single version of that.

I'm also going to pat myself on the back for getting the circuit right without breadboarding it first, and getting the etch right with that press-n-peel stuff (and a sharpie). Next up is a Bass BB I etched alongside it, then an Arduino switch (that's gonna be fun - 9 loops, true bypass, programmable with LCD - coded and ready for breadboarding!). YAY!

Thanks for the insight, guys!

[madbean]

What Jon suggested - make R9 a 1k5 or 2k. You can also increase C11 to 68n. Maybe not both mods though - that might make it too dark!

Another cool trick is one Forrest came up with: instead of tying pin1 of the Delay control to ground, do it on pin2. Now connect pin1 to a large cap (like another 100n) and put in parallel with the existing C9 cap. As you turn the delay pot all the way up, the new cap goes to ground and increases your filtering. He's a smart one.

[culturejam]

Another cool trick is one Forrest came up with: instead of tying pin1 of the Delay control to ground, do it on pin2. Now connect pin1 to a large cap (like another 100n) and put in parallel with the existing C9 cap. As you turn the delay pot all the way up, the new cap goes to ground and increases your filtering. He's a smart one.

I wish I could take full credit for that one. It was actually Alex Shroyer (earthtonesaudio) that came up with that one while helping me out on something delay-related. I just tacked it on to the simple delay circuit I was working on. 

[alanp]

Despite what Alan says, the current version of the CD has about the same delay path stuff as the DBD or most others, it just has a greatly simplified dry path. The older CD was pretty much "live with it."

Sorry, I hadn't realised that. My CD is the older one, and I hadn't realised at all that the new one had altered the filtering

[greysun]

In case anyone is wondering what it looks like under the hood...



I had a ton of those huge .1u caps and wanted to use them (tossing them is wasteful). I also wanted to test out using press-n-peel for an arduino switch, so this is the result. (I also etched and built a bass BB, but I'll post that in another area).

I still had to use low profile jacks and set it up kinda like a 1590a, but in a 1590b. Either way, cool to do it one sided with no jumpers. Teacher loved it, and that's all I can ask for. :-) said it would be good for his rockabilly class.

If I build another one for myself (which could be likely), I'll order it from madbean. This was fun, but ultimately a pain, and I'd rather the smaller enclosure.

[diablochris6]

Looks good. I applaud you for designing a layout so that you don't waste parts.