Dimension P - New PT2399 Chorus

Started by CynicalMan, October 29, 2011, 09:42:00 PM

Previous topic - Next topic


Quote from: frequencycentral on November 17, 2011, 12:37:31 PM
Just a couple of thoughts to ponder on multiple PT2399 designs:

- it would appear from recent reseach that pins 3 and 4 are connected internally by a low value resistor. If you look at the example circuits on the PT2399 datasheet, the delay pot is connected to digital ground. So maybe connect the anti lock up circuit should be referenced to pin 4 rather than to ground.

I tried that and I didn't hear any improvement. Merlin tested using pin 4 instead of ground in a normal PT2399 circuit and he didn't see any noise improvement. It could be one of those weird variances between chips. Anyway, it works with that fix, but I left it out because I thought layouts would be easier with the anti-lockup going to ground. You can do it either way.

Quote from: frequencycentral on November 17, 2011, 12:37:31 PM
- it would further appear from recent reseach that the PT2399 dumps a load of hash into the 5v supply rail, so the 5v supply rail should not really be used as a convenient vref for other stuff (not that Alex has done this). So, as the PT2399 is dumping hash - would it not make sense in a multiple PT2399 delay to use a seperate 7805 for each PT2399, thus further isolating and reducing hash?

I didn't know this, but I'll look into filtering. Hopefully I can get improvements without using another 7805. I am using the 5V line with the LFO inverter, but that's not in the signal path so it shouldn't add noise (I think).

I'm also looking at the pre-emphasis thread, so I'll run some real noise tests tomorrow and try these out..


I'm using a real Vr (as per your original schematic) to feed the inverter, my buffer stage and mixing stage.

Schematic is done...box art is done....layout......(not quite there yet  :icon_redface:)

I've added an input buffer, a mixing stage that will receive (Dry, Voice1 and Voice2) also adding a 160Ω in parallele with the 270Ω via a switch to bring it to 100Ω to get some TZF.

All this will fit in my 125B.


This layout is unverified so let me know if you try it.  I'll try to etch the board over the weekend and so how it goes.  Based on CynicalMan's 1.1 version.

Bust the busters
Screw the feeders
Make the healers feel the way I feel...


Thanks for that Gordo.

If you don't mind.....I'll use your layout as inspiration for my incarnation of this project.


My pleasure, I'll post the Eagles if you'd like.

Here's the ground pour version of the board
Bust the busters
Screw the feeders
Make the healers feel the way I feel...


Nice layouts! Good luck with the PCB.

I tested the 5V supply noise, and there wasn't any more with two PT2399s than there was with only one. There was very little high frequency noise in general, much more 60Hz and LFO noise. Which isn't to say there was a lot of those, but the digital noise was much smaller still. It was only 3dB higher than the base noise of my audio interface.

Has anyone built or breadboarded this yet? I'm curious about your thoughts on the sound.


I'm at attempt #6 of a PCB layout of my incarnation of this. :icon_redface: :'(
Attempt #7 might be the good one.

Can't wait to build this!


OK PCB done (I needed to fits everything on a 2" X 3" board), etched, drilled, thinned.......now need to populate it.

Can't wait to ear it.


Wow, not bad.
Let us know how it turns out!



Board populated......will finish this during my Christmas holidays (hopefully)

Redoing the bedroom......not much free time....

No worries I will post my progresses   :icon_biggrin:


Using the stereo version and adding a dpdt (on/off/on) after R17/R23 (jumper switch contacts 1&3 on the R17 side) and changing the mono/stereo switch to (on/off/on) would give you the best of both versions. Loads of output configurations possible with only 2 switches.

In Mono mode you have:
2 voice + dry at out A (dry only at out B)
2 voice wet only at out A (nothing at out B)
2 voice wet only at out A (dry only at out B)

With mono/stereo in the middle (off) position:
1 voice + dry at out A (dry only at out B)
1 voice wet only at out A (nothing at out B)
1 voice wet only at out A (dry only at out B)

In Stereo mode you have
2 Voice + dry in stereo using both outs (or 1 voice + dry if only using one output)
2 Voice wet only in stereo using both outs (or 1 voice wet only if only using one output)
2 Voice wet only in stereo at both outs with dry at out B only.

Hope that all makes sense. Not sure how useful the last one is?

I will be knocking up a vero layout of this quite soon.


That's neat. I figured that there were 6 'useful' positions myself:

Both voices with Dry
1 voice with Dry
Both voices without Dry
1 voices without Dry
Stereo with Dry
Stereo without Dry

But I couldn't find a way to do it with two switches. I was planning to use three on/on switches with redundant positions, but your system is better.

@ Alain,
I just noticed a potential issue with your setup. If you use pots instead of switches, that could reduce the cutoff of the output low pass filter. You might want to reduce the value of the 4.7n capacitors or omit them altogether if you haven't already.



Just finished putting this vero layout together. Based on the Stereo schematic with my "best of both" switching from a few posts above.
Component labels follow the schematic. Due to the software the component list is not in order and R1=1M has dropped off the bottom.
Havn't had a chance to check over it so there may be some mistakes. I will update this post for any corrections and verifications.

note: one or both of the pots function may be reversed.

A better view of the cuts and jumpers


Thanks...  :icon_redface:
Noticed a few errors that I'm going over at the moment. I'll update a bit later.

edit: Fixed a few mistakes, layout above now updated.


Just finished populating my vero. Gets pretty tight but do-able as long as your caps aren't too big.
The board itself will fit in a 1590b but I'm not going to attempt to squeeze sockets, pots and switches in with it.

Hopefully, I'll get the wiring, pots and switches on tomorrow and see if it works.

Does anyone know how much current is drawn from the 5v regulator? I wouldn't mind using the smaller TO-92 package but the ones I have are max 140mA.


That's a neat vero! Every time I look at a DIYLC vero layout, those things look horribly messy and a waste of space; but, seeing how tight they turn out in practice, I might just save myself etching a PCB and building this on vero... looks great!


I took a quick look at your vero, and I did notice a couple of minor errors. I'll try to take a closer look tonight.
The input of Delay 2 is connected directly to the input instead of after the input capacitor. The quick fix would be to replace the jumper from D20 to K20 with a 0.1uF capacitor.
I'm not sure which side of the pots the layout is supposed to show, but make sure to use pins 2 and 3 on both pots.

I'm not sure about the current draw. The regulator does get warm. I can't see it using 140mA but I could be wrong.

Otherwise, it looks great!