pt 2399 echo chip equivalent

[Frances Rhodes]

hi everyone

(i'm reposting the topic here since i'm not sure where to put it)

i'm looking for a delay chip (digital or analog) that would allow a time delay of less than 5 ms, maybe even less than 1 ms.
i read the pt 2399 datasheet, it looks really easy to use but it seems that le smaller time delay possible is about 30 ms with an external resistance of 0.5 ohms.
would anybody know of a chip that may allow really short time delays and not be a pain to use... and not be a micro controller either?

thanks

[Mark Hammer]

Simplest thing is going to be a low-capacity BBD, really.  A relatively simple thing to clock a 1024-stage chip like an MN3207 equivalent.  And since you are aiming for a nice reasonable delay time (not too long, not too short)  the clocking would be simple and the filtering required fairly minimal.   Pushing a 1024-stage BBD lower than 1ms might get a little tricky and require buffering the clock, but certainly 1ms is well within reason.  I would suggest a lower capacity (256 or 512 stages) chip for shorter delays, but they're tough to find and no longer in production.

[Earthscum]

I've been looking for something like this as well, though not intensively. I'm confused... I'm looking at the CoolAudio sheets (I have a pair of V3207's and a pair of V3102 clocks). The 3102 doesn't list a max frequency, nor a range for that matter (I think I had to look up the MN datasheet before). The 3207 has listed a range from 10k-200kHz, delay ranging from 2.5mS (I calculated 5mS, apparently my math is off) to 51.2mS. Can these things really be overclocked to (500kHz?) or is there a different method?

[Frances Rhodes]

hi

thanks for the reply!

i searched on google a little and found digital delay chips (the NJU 9702) used for surround effects that go down to 0.5 ms, and like the pt2399 with integrated VCO for clock signal, and also ADC and DAC, but the delay time is not continuously adjustable, it's only in 0.5 ms steps.

i have absolutely no skills in programming chips, but couldn't this be written on a simple PIC chip (like a DIP-16 or something like that, since it's such a short delay time, it wouldn't need a lot of memory to be loaded) or do short delay times require higher computer power to be achieved and thus would need a real micro controller?

[Hides-His-Eyes]

dsPIC would be more suitable, but I'm not saying it's not possible.

an AVR would probably be better than a PIC too

[Mark Hammer]

A 3207/3102 combination can be used to generate a 1-5ms delay easily.  Otherwise we would not see it in so many budget flangers and choruses.  And if it wasn't in so many budget choruses and flangers, we wouldn't see it in production at all.

[Frances Rhodes]

ok thanks!

i'll try to find AVR dedicated forums and ask there if someone knows about this.
i'll keep the 256-stage bbd (i also found about 128-stage bbd too, but probably much harder to find) option in mind, but an analog delay line seems pretty hard for me, since i'm still a beginner. i'll keep searching for a pt2399 alike.

[Fender3D]

I'm confused... I'm looking at the CoolAudio sheets...

CoolAudio datasheets simply suck, they're an incomplete copy of the older MNs datasheets
Let's hope the chips aren't an incomplete copy too...  

@Frances
your quest depends on what are you willing to build.
If your delay will remain static you might as well give a shot at the NJU stuff.
If you need a continuously modulated delay, a BBD will do the job better (without "steps"), with less parts and with an easier layout

[Frances Rhodes]

@Frances
your quest depends on what are you willing to build.
If your delay will remain static you might as well give a shot at the NJU stuff.
If you need a continuously modulated delay, a BBD will do the job better (without "steps"), with less parts and with an easier layout

what i'd like is some sort of a regular delay effect but with super short delay time. i have a boss DD-20, which is supposed to go down to 1 ms in 1 ms step but i doubt the clock is really precise at those extreme settings. so i want to build a device with delay time pot (for continuous variation), feedback pot and probably an intensity pot too. since it's not for chorus/flange effect, i don't need to add delay time modulation.
if bbd chips can be overclocked, then the mn3206 should be able to provide less than 0.3 ms of delay time (as mentioned on the datasheet), this may suit my needs. but i'm afraid analog chips and circuit may lead to signal losses, specially for mid/high frequencies.

i studied the deluxe memory man circuit before (the one with only 2 mn3005). i'm not sure i understand every part of the circuit, but i guess this could be kind of a model to start from.
only 1 bbd chip, maybe no need for a 570/571 (used for compressor, i think?)... i could use a 555 clock osc for a simpler circuit, would it induce more noise than a fancier oscillator?

[Fender3D]

You say it's not for chorus or flanger effect...
I assume you know what a heavy feedback will do with 1ms or less delay.
Tomas (Thomeeque) did a huge job with his DMM retrofit.
It uses a MN3207, so plenty available (its just a matter of trying with V3207), you simply have to delete the LFO part if unnecessary, and enjoy the filter matrix setting.

[Frances Rhodes]

You say it's not for chorus or flanger effect...
I assume you know what a heavy feedback will do with 1ms or less delay.
Tomas (Thomeeque) did a huge job with his DMM retrofit.
It uses a MN3207, so plenty available (its just a matter of trying with V3207), you simply have to delete the LFO part if unnecessary, and enjoy the filter matrix setting.

yes i know. i actually want to use it as a note generator with white noise on input exactly like Jean Claude Risset did with "Sud" with the sound of the ocean that he recorded. except that he used computers, and i want it to be the size of a pedal (if you don't know about it, it's pretty hard to explain it in english since i'm french, but once the "song" was recorded, it was intended to stay "freezed". when it was played in front of an audience, there was a tape recorder on stage, someone came to turn it on, and off at then end of the show. it was the first time recorded music was played in front of an audience without any other "performer", it was a really interesting work! but anyways)
i haven't heard about Thomeeque's job on the DMM, i'll go look for it, thanks a lot for the information!

[Scruffie]

You say it's not for chorus or flanger effect...
I assume you know what a heavy feedback will do with 1ms or less delay.
Tomas (Thomeeque) did a huge job with his DMM retrofit.
It uses a MN3207, so plenty available (its just a matter of trying with V3207), you simply have to delete the LFO part if unnecessary, and enjoy the filter matrix setting.

yes i know. i actually want to use it as a note generator with white noise on input exactly like Jean Claude Risset did with "Sud" with the sound of the ocean that he recorded. except that he used computers, and i want it to be the size of a pedal (if you don't know about it, it's pretty hard to explain it in english since i'm french, but once the "song" was recorded, it was intended to stay "freezed". when it was played in front of an audience, there was a tape recorder on stage, someone came to turn it on, and off at then end of the show. it was the first time recorded music was played in front of an audience without any other "performer", it was a really interesting work! but anyways)
i haven't heard about Thomeeque's job on the DMM, i'll go look for it, thanks a lot for the information!

I think he means Electric Mistress not DMM...

You could also just use a simple 4047 Clock with 4049 Buffer, in fact, Thomeeques Through Zero Flanging board could be perfect.

I've got a 3206 for sale at the moment though if you want it, but a V3207 will do just as nicely if buffered.

[Frances Rhodes]

You could also just use a simple 4047 Clock with 4049 Buffer, in fact, Thomeeques Through Zero Flanging board could be perfect.

you mean use the 6 different stages of a 4049 in series and feed the supply voltage pin with the clock signal?

[Fender3D]

I think he means Electric Mistress not DMM...

Yep 
DMM, DEM, EM...
EHX BTW 

[Frances Rhodes]

You say it's not for chorus or flanger effect...
I assume you know what a heavy feedback will do with 1ms or less delay.

thinking about it again, i guess you were talking about a reverb effect, but that's not what i'm looking to build. i'm looking for 1ms or less delay times because i tried this (what i explained earlier, generating notes by delaying an input signal, instrument, or even white noise) with my Boss DD-20 and it worked with the shortest time available, which is supposed to be 1ms, but the clock of this pedal is really not precise and i can't measure the delay time exactly, that's why i said i was looking for this range, but it may not be it.

you mean use the 6 different stages of a 4049 in series and feed the supply voltage pin with the clock signal?

ok, i was confused at first but i get it now, sorry for the newb stupid question!!

[Fender3D]

I talk about heavy feedback 'cause you said you want a feedback control..
Anyway
for short delays you might check EHX "full double tracking effect" or whatever is called, subbing MN3005 with a shorter BBD (MN3008 is a pin to pin replacement, MN3007 will need a daughter board or another PCB)

[cctsim]

This ES56028 chip can do in "surround mode"  4.1 ms of delay. Datasheet link below:

http://www.cbtricks.com/radios/galaxy/datasheets/ic/pdf/es56028.pdf

It is used in the digital versions of Boss CE-5 and CH-1, and in the Danelectro cool cat chorus (probably a copy of CH-1).

From the schematic it looks like very similar to the PT2399.

[frequencycentral]

This ES56028 chip can do in "surround mode"  4.1 ms of delay. Datasheet link below:

http://www.cbtricks.com/radios/galaxy/datasheets/ic/pdf/es56028.pdf

It is used in the digital versions of Boss CE-5 and CH-1, and in the Danelectro cool cat chorus (probably a copy of CH-1).

From the schematic it looks like very similar to the PT2399.

Nice find! Super simple flanger on the way!

[Frances Rhodes]

This ES56028 chip can do in "surround mode"  4.1 ms of delay. Datasheet link below:

http://www.cbtricks.com/radios/galaxy/datasheets/ic/pdf/es56028.pdf

It is used in the digital versions of Boss CE-5 and CH-1, and in the Danelectro cool cat chorus (probably a copy of CH-1).

From the schematic it looks like very similar to the PT2399.

thanks a lot, i'll go read this datasheet asap!

now this brings me question, what is the difference between chorus and flanger? is it the delay between dry and wet that is shorter for flanger effects or is it something else? i know they both work the same, a modulation of the delay time, but that's all i know

[frequencycentral]

It has a little brother too:

http://ic.wwbj.net/pdf/ES56033.pdf

......though with no Rosc/Fosc delay time table, and no stated delay times on the data sheet, it's anyone's guess as to how low you can go.

this brings me question, what is the difference between chorus and flanger? is it the delay betweek dry and wet that is shorter for flanger effects...?

Yes, exactly that. That and a feedback path.