Posted: Wed Jun 01, 2016 4:00 am Post subject:
reverb, echo, flanger, chorus etc... Subject description: Kindly asking input...
So I have this large spring reverb tank of about 40 in length.
I have multiple PT2399 and have ordered some MN3102 MN3207 combo's.
What the spring reverb will do is obvious.
I plan (and have started building) to use three PT2399 on one board.
A PT2399 is an echo audio processor ic DATASHEET
The 1st will be controlled from ~40mSec - 151mSec, the 2nd from 151mSec - 273mSec and the 3rd from 273mSec - 350mSec.
They will be placed in serial, with a potmeter/tap to the input so I have a delay from 40mSec up to a max of ~800mSec.
I will use a CV for the delay settings.
The MN3207 MN3102 combo (DATASHEET MN2307) is a BBD (MN3207) and its clock generator (MN3102) (DATASHEET).
The MN3207 is a 1024 stage Buckett Brigade Device, which can produce delays from 2,56mSec - ~51.2mSec and the plan is to have several of these on one board.
So I think these three different approaches covers all the delay times feasible.
Maybe it's a good idea to put these all in one 19" rack than the reverb tank will fit also.
Now I also have to figure out how to produce effects like flanging and chorus and if this is at all possible with this (still imaginary) setup?
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
Posted: Thu Jun 02, 2016 12:15 am Post subject:
Flanging uses a short delay (PT2399 type delay chips are too long) of varying time length (typically a cheap two IC triangle LFO), which is then mixed back in with the dry signal. The ADA Flanger is one of the most highly regarded. The EHX Electric Mistress is one of the most common.
Chorus uses a slightly longer delay (but slightly too short for PT2399, although there are some hacks that pull it off), with varying time length (same LFO, although with less depth than flanging).
With a good flanger, you can cop chorus sounds. But it's far less common to be able to do flanging with a chorus. Typically, short BBD delays (like the 3207 and 3007) are used.
So for flanging and chorus I can use the MN3102 MN3207 combo with a triangle input the cv?
The PT2399's will be used for echo and reverb and the spring tank for reverb only.
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
Posted: Thu Jun 02, 2016 2:09 am Post subject:
I strongly suggest checking out pedal flanger schematics. ADA and EHX used their own clock source, NOT the MN3102 chip. (A 4000 PLL chip's VCO for the EHX, something a bit more complicated in the ADA from memory.) Lovetone used the MN3102, but with a LDR attached to provide the CV from the triangle LFO.
For the delay time settings of the PT2399's I use an AD5144 which is a quad digital controlled potmeter and thought to use it as well for the MN3102 to set the delay time.
There will also be an Arduino (clone) involved to measure Control Voltages and program the potmeters.
I'm still pondering about it all, but are already building the board with the Arduino, an AD5144 and three PT2399's.
For the chorus, how many delays do I need for a nice effect?
Short film using one PT2399, one channel of the AD5144 digital potmeter and one Arduino clone.
I start with the dry signal and than add reverb.
Very weird effects, by just changing the cv of the delay time.
This audiofile starts with the dry signal and later the feedback is turned up.
I'm using my sequencer for a nice loop.
The delay time is modulated with a ramp voltage with varying frequency (up to about 500Hz is possible, after that you get a major aliasing.
When listening to this file, there is a lot of background noise, caused by bad grounding and lack of capacitors across Vcc and Gnd.
The next file sounds better, I promise
reverb-01.mp3
Description:
starts with the dry signal and later the feedback is turned up.
phasing/flanging is possible with PT2399: use 2 of them in parallel : 1 with a fixed delay of 35ms for example, the other freq.modulated around 35ms,
(e.g. 30..40ms) then just subtract the outputs.
For me it sounds strange to combine a 60ct. PT2399 with a 5,- digital pot and an even more expensive BBD chip. I think the noise you mentioned in your mp3 example can't be improved very much, the PT2399 datasheet just lies about S/N-ratio. It's just what you get for the 60 cent. I used a simple 1 tran. current sink to control delay time, which is good enough for what you can get out of the PT.
the noise has been drastically minimalised by changing the ground connections, so little problems here.
the reason why I use the dig pot is because I got them for free and its a great way to control the delay time.
I bought the bbd +clock generator at AlieExpress, 5 pcs for $2.19
the bbd is meant to be used for short delay's in parallel, and yes, here the delay will al so be controlled by one AD5144 for flanging sounds
besides that: what do you think about the sound? remember: this is one PT2399, and the finished module will have 3, in series or in parallel, to choose with switches
good, that you could reduce the noise. My experience is, at least for longer delays, that you have to live with -50..60dB noise and never the > -80db from the datasheet.
free digipots, great, nothing can beat this
the price for bbd+clock sounds very good. I pay 2.50 for each. here in Germany.
" remember: this is one PT2399, and the finished module will have 3, in series or in parallel, to choose with switches"
yes, that's why I suggested the phasing solution with 2 (ore more) PTs
(if you can live with the 35ms dead time)
well free... it is called samples, this is how i get really nice stuff for free from Analog Devices, Maxim and Texas Instruments.
it's part of my job to experiment with new circuits, and this way it is also benificial for my hobby
the noise is most prominent when using long delays, for this reason I use three PT2399's, where the longest delay will be ~300mSec, when connected in series the total delay will be ~1 sec. max.
but with the addition of the possibility for each PT2399 to feedback to the chip itself or to the main input. my guess is that this will give nice reverb/echo signals.
It is true, when I use longer delay times, the noise is more prominent, especially in lower longer notes. There you can hear the steps of the ADC/memory/DAC.
But this demo is (just like the previous one) directly from the output of the PT2399, without any filtering!
Just looked at the Abductor schematics, and they (four of'em) use all the same delay, which is modulated with an LFO.
My design will have (and already does) have CV possibilities.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum
Forum index » DIY Hardware and Software » Developers' Corner
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
