Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[SineHacker]

Been working on this over the weekend, gonna box up a prototype in the week and do a short production run after that, enjoy!

Homemade Delay

p.s I saw there was an old thread on here that was pt2399 delay related - I think it was discussing the old getlofi kit. When I was looking at the datasheet I was also confused by AGND and DGND as the sheet doesn't offer much explanation on what the difference was - so I did some research online and people generally conclude by saying "join the two separate grounds at one point" - but I don't think that is the case here, in fact I'm pretty sure; I think the chip has AGND and DGND so you can connect all AGND routes to that pin, and the chip will then filter it over to DGND - you can then connect DGND directly to your GND line - doing it this way got me far far better results than connecting the two lines at any point - I don't know if I should post this over in that other thread cuz it's probably really old

[Circuitbenders]

this is pretty much everything you need to know about AGND and DGND, but i assume you must have read it already if you've researched this: http://www.analog.com/library/analogDialogue/Anniversary/12.html

According to the datasheet if you don't actually have any extra digital circuitry i'd be inclined to wire the AGND and DGND together at the IC, but i could be wrong about that. I'm not sure what you mean by ' the chip will then filter it over to DGND'?

If you're connecting your ' DGND directly to your GND line' then you are already 'connecting the two lines at any point' aren't you?

Try building a delay with the PT2395. You can do some excellent stuff clocking it with an LTC1799
http://www.circuitbenders.co.uk/newsarchive/danecho.html

[SineHacker]

Yeah I had been through that page, it was helpful - and I did try tying it at the pins labelled AGND and DGND on the PT2399 as well at some other points I had seen recommended from some other ancient forum posts - but it definitely has less noise and operates better if they are unconnected and you route everything the grounds to AGND in the schematic to that pin - I don't know how to describe it properly but I think the IC deals with it internally (as a guess). Hey its working for me anyway!

Yeah I have seen your pt2395 mods it was me asking on Facebook if you had any left knocking around I haven't managed to get hold of any of the chips yet, I haven't seen them on EBay at all

[SineHacker]

Oh and I just meant that DGND is connected to ground that goes back to the battery and regulator ground AGND doesn't connect to this at all. I'm pretty sure the engineers page states that every IC can handle it differently - but I haven't looked at it in a long time. I don't want anyone getting misinformation here! Just saying this worked really well for me in this situation, but I couldn't explain why! 

[Circuitbenders]

I'm not sure i'm understanding you properly, how is your AGND actually grounded if it doesn't connect to the regulator ground or the battery negative in attempt to not connect it to the DGND?
What are you amplifying the output through? I would imagine you might have problems connecting it to external devices, but if it works then it works i guess.

The way i read it you should connect the ground on any digital stuff to the DGND to prevent it introducing noise onto the analogue ground. If you haven't got any other digital stuff then i don't suppose separating the grounds matters that much.

utsource will sell you PT2395's for about $8 each, but i think you need to buy a RAM chip as well http://www.utsource.net/

[SineHacker]

I think (with emphasis on the think!) that AGND is connected to DGND internally in the PT2399. Could I possibly take a peak at the datasheet you have, the one I downloaded doesn't mention anything about the two grounds - I am grounding the input and output jacks and the time resistor with the battery ground and also the DGND pin on the chip (pin 4 i think) Everything that the schematic shows going to AGND I have just connected to pin 3 - not using a jumper anywhere to link AGND to DGND. I'm pretty sure that if I use a jumper across pin 3 - 4 then I get a massive increase in noise and a big decrease in the length of echo feedback - I will double check this today though and report back  

[SineHacker]

Well, I plugged the circuit into a nasty noisey old amplifier and it worked exactly the same whether I tied AGND to DGND or not - so I guess I will leave it tied just to be on the safe side

[SineHacker]

I did a couple of prototypes and ended up with a design I'm pretty happy with:

[soundcloud]http://soundcloud.com/aidantaylor/jellyverb-demo[/soundcloud]

[SineHacker]

some nice pics here:

http://soundofginko.com/monthly-drone-july-jellyverb-complete/

[Circuitbenders]

Nice 

This is kind of off topic, but does anyone know if UV LED's are actually Ultraviolet or just purple? I had a some years ago that were definitely UV and had a darker lens, but every time i've had any with clear lens' since then they have been pretty poor in comparison

[Gordonjcp]

They are UV, just shine them at a high-vis jacket...

[SineHacker]

These one are incredibly bright in comparison to the blacklight (which is more effective!) but they do work - I was trying to balance illuminating the image with not effecting the LDR - they don't effect it too much anyway unless you shine them directly towards its top side. I was hoping they would be a little less directional - will experiment some more next time.

I am wondering if it is possible to get translucent stickers with a solid black image along with invisible UV highlights image (like the hand drawn designs), i've seen translucent UV security stickers so I guess it must be possible, any thoughts?