Origins of the PT2399

[Strategy]

I'm finishing my girlfriend's x-mouse/chanukahmas present - a Magnus Modulus configured for desktop/synth/vocals use.

As I'm doing this i'm wondering about the PT2399's origins, supposedly for Karaoke machines.

What the heck is a chip this crazy *doing* in a Karaoke device? Not doing conventional delay, I would assume? Anyone know how those are used in those machines?

- Strategy

[slacker]

Most of the applications listed in the data sheet relate to using it as a surround sound processor, ie just to create a very short delay. I think in Karaoke machines it's used as an delay the same as in a pedal, someone here, maybe Top Top, gutted one and that's how it was set up.

Most of the crazy stuff has been done by us making the chip do things it's not really designed to.

[Maudibe]

And this is about the only time a guitarist says 'thank god for karaoke' 

[Mark Hammer]

Large capacity BBDs actually found greater use, world-wide, in karaoke machines than in anything related to guitars (the DMM notwithstanding).  The Princeton chips (just like the Holtek and the Mistsubishi) simply carry on that tradition into the digital domain.  Indeed, it was probably the desire to have a chip that conveniently provided echo without the hassle and parts cost of analog chips that spurred development of the single-chip digital solutions.

Now, as to why karaoke machines invariably come with echo, that's another matter.  Personally, I have yet to hear any popular music from South Asia that does NOT have some sort of echo on it, though I am less familiar with East Asian pop music, and with whether it too comes with echo all the time.  Of course, whether it does or doesn't, adding some echo to a singer nearly always makes them sound more grand, so I imagine echo chips find a spot in karaoke machines as a sort of "talent" button.  As well, karaoke machines can often find themselves in contexts where the room acoustics are not going to make you sound particularly huge.  And finally, reverb and echo may well be part of the backing track, so the option to have the voice ambiance match that of the backing track is something you want to be able to provide so that it appears to fit the song.

[askwho69]

Well in my knowledge.. karaokes'are invented by filipinos because Filipino people lovesto sing in karaoke and thats how bon jovi, heart and other popular singers known in our country ehehehehe

[ExpAnonColin]

Large capacity BBDs actually found greater use, world-wide, in karaoke machines than in anything related to guitars (the DMM notwithstanding).

I think they were rather widely used in old TV sets as a fake sort of reverb as well.

In terms of karaoke machines, I've found many echo chips unknown to the guitar pedal world by taking apart karaoke machines.  The market for karaoke machines is just so many orders of magnitude bigger than that for guitar pedals, so even if a small fraction include an echo feature, it's still going to be a waaayy bigger market.

-Colin

[Mark Hammer]

I gather those various Mitsubishi (NOT the same as Matsush*ta) chips like those below are some of the ones you're talking about.
http://www.alldatasheet.com/datasheet-pdf/pdf/86751/MITSUBISHI/M50195P.html
http://www.alldatasheet.com/datasheet-pdf/pdf/1414/MITSUBISHI/M65831AP.html
http://doc.chipfind.ru/pdf/mitsubishi/m65845afp.pdf

[Top Top]

Most of the applications listed in the data sheet relate to using it as a surround sound processor, ie just to create a very short delay. I think in Karaoke machines it's used as an delay the same as in a pedal, someone here, maybe Top Top, gutted one and that's how it was set up.

Most of the crazy stuff has been done by us making the chip do things it's not really designed to.

I gutted a karaoke boombox to make into a delay but it actually had a different echo chip I think... can't remember now though.

I did just open a karaoke mic with built in echo though and it had a little SMD 2399 in it. Both were set up in a fixed length echo type of configuration. Echo helps mask bad singing and playing. As a bad musician, this is something I know well

[Strategy]

This is great info. Thanks for entertaining my early morning wonderings.

I like the 'misappropriated' use of pt2399 as a guitar effect most! It strangely puts this chip in the same genre as video game chips, Commodore SID chips, and "robot voice toy" chips in terms of its origins in "re purposing"...for the greater good of pedal freaks everywhere.

strategy

[askwho69]

I just made robote delay and i kind a remember the sound of karaoke machine echo! LOL!

[faptdivers]

say vvhat princess?

[chi_boy]

I'm not sure what part of this thread sparked this thought, but would it be possible to use the PT2399 in surround/delay mode and separate the raw and delayed signals to stereo outputs?  The thought would be to have a splitter going to 2 different amps, but 1 amp only sees the delayed part, or just the single repeat, and the other sees the unaffected signal.

With stereo outs, it could be interesting to have effects after the delay split too.  Hmmm.....  if only my bread boards weren't jammed with unfinished projects...

[Hides-His-Eyes]

chi boy: yes, that would be trivial.

OK guys, I'm finding the bit of PT2399 circuits that comes out the output a little impossible to parse. I realise that low noise requires low pass filters, but I know what low pass filters look like... Is there something I'm missing? It just seems like a weird setup.

EG echobase:

http://www.aronnelson.com/gallery/main.php/v/slackers-stuff/album170/echobase.png.html

100n across the op-amp, low pass. 10k and 15n, low pass. 1n across the next op-amp, low pass. What are the 10k and 47k doing there? 10k and 15n, low pass filter...

Does this thing not end up ridiculously dark? Or are we just putting a 'cliff' at 1kHz?

What are the 10k and 47k doing in that arrangement on the input?

[slacker]

100n across the op-amp,

The pins with the 100n across them are integrators that convert the digital signal back to analogue.

low pass. 10k and 15n, low pass. 1n across the next op-amp, low pass. What are the 10k and 47k doing there?

These all form a multiple feedback lowpass filter. There's some info here http://sound.westhost.com/articles/dwopa3.htm#filt. The input is the same with the 20k resistor and 4n7 cap being part of the filter. The basic arrangement is from the datasheet with values tweaked to taste. The resistors were chosen to give some gain, the output of the PT2399 needs to be louder than the input in order for the feedback to work.

Or are we just putting a 'cliff' at 1kHz?

Yeah that's it, the input filter is somewhere around 4kHz and the output one around 1kHz.

For what Chi boy wants he can probably just use the example in the datasheet.

[chi_boy]

For what Chi boy wants he can probably just use the example in the datasheet.

I guess is is a simple task after all.  I didn't see it on the datasheet, but the Rebote 2.0 shows exactly how to do it.  Now to marry it to the hum free ABY at GGG.