Author Topic: MN3007 ADA Flanger Clone Questions (Read 224771 times)

njkmonty · « **Reply #300 on:** July 09, 2011, 07:26:01 AM »

and some voltages.
from boss 9v power supply = 8.79v
from charge pump = 16.50v
after regulator = 14.17v

4007 4047 3007 4049
1- 0.34 1- 12.2 - 9.1 1- 13.93 Pins 1 & 16- 14.25
2- 0.34 2- 0.7 - 3.9 2- ~6.3 Pin 8- 0
3- 0 3- 7.4 3- 7.99 Pin 13- NC
4- 0.04 4- 14.25 4- 0.87 ALL other pins- ~6.37
5- 0.04 5- 14.25 5- 0
6- 0 6- 14.25 6- ~6.26
7- 0 7- 0 7- 4.9 - 6.0
8- 0.05 8- 0 8- 4.9- 6.0
9- 0.37 9- 0
10- 5.1 - 3.7 10- 5.9- 7.2
11- 7.2 - 7.8 11- 5.9-7.2
12- 0.7 - 3.2 12- 0
13- 14.25 13- 12.3 - 9.2
14- 14.25 14- 14.1-14.5

IC 1 IC2 IC3 IC4 (LFO; slowest speed setting)
1- 7 1- 4.7 1- 6.9 1- 13.1
2- 7 2- 4.7 2- 5.6 2- 7.31
3- 3.4 3- 4.6 3- 6.5 3- 3.6 starts at this and goes up slowly
4- 14 4- 14.1 4- 14.9 4- 0
5- 7.4 5- 7 5- 7 5- 7.0
6- 8.0 6- 7 6- 7 6- 7.0
7- 8.0 7- 6 7- 7 7- 5.5 starts to drop slowly
8- 7 8- 7 8- 5.0 - 3.6 8-13.9
9- 7 9- 7 9- 5
10- 7 10- 7 10- 4.4
11- 0 11- 0 11- 0
12- 0.01 12- 6.9 12- 3.8
13- 12.87 13- 7 13- 3.8
14- 12.87 14- 7 14- 6.09 - 7.1

some major differences i noticed to others are
pins 12,13,14 ic1

oldschoolanalog · « **Reply #301 on:** July 09, 2011, 10:27:35 AM »

There should be 15V after the regulator. No exceptions. There is not enough V going into it (check the datasheet of your reg for dropout V). This is one of the reasons I like to use an 18V PS and no charge pump (among others). If you are using a MAX1044 there will not be enough current for this project. Can you bypass the charge pump and get 18VDC to the input of the reg? Try that and then check your V's.
I'm at work. More thoughts later.

njkmonty · « **Reply #302 on:** July 09, 2011, 07:50:25 PM »

do you think that slight drop in v's will make a huge difference?
ive seen a few 18v power supplies around, ive noticed 2 types
both dc but one is transformer type and the other is not.
since the 15 volt regulator will be doing its thing, does it matter in what type of 18v dc supply i use?

oldschoolanalog · « **Reply #303 on:** July 09, 2011, 09:23:42 PM »

First go to page 11 of this thread. I posted the V's of a properly functioning unit there.
Next read through this entire thread. It's a bit mind numbing, but not as much as the original "mega A/DA thread". Brew some strong coffee/tea and have at it.
My personal choice is to go w/the old fashioned transformer type PS. I have never had a problem with those; innefficient as they are & all that. There may/may not be any issues w/a switched-mode PS. I don't use them because they can potentially introduce heterodyning (maybe not). Just make sure it can supply ~ 100 mA @ 18VDC. If you read through this thread you will see several people have had issues w/the charge pump setup. Some have not. Your choice.
All the Best!

Scruffie · « **Reply #304 on:** July 12, 2011, 06:35:37 AM »

Have you got the diode in after the charge pump? Dropping that might help...

But I had a slightly lower than 15V reading and my A/DA is fine, I wouldn't worry about 1V of loss personally.

njkmonty · « **Reply #305 on:** July 13, 2011, 01:46:26 AM »

yes ive got the full charge pump circuit, in from the latest schematic literature.including the diode 11 (which is supposed to be labelled as diode 8 or 9?!!!)
i bought today a dick smith (aussie version of radio shack) multi dc powersupply, its obviously not a transformer based one, but i can take it back if i dont need it.
it goes from 12v, 15v, 18v, and 24v. the voltages measured from it where pretty close to advertised.
what i noticed was using the 15v regulator, would loose about .2 of a volt. ie 15v in , 14.8v out.
this occured with either the 18v or 24v. after 15v regulator the circuit was getting around 14.4v.
now when i used the 15v selection from the power supply, and fully omitting the charge pump/15v regulator
the measured voltage for the circuit varied from 14.85 - 15.05volts!
surely this will be good enough?

I will try and do another voltage chart testing thingy,
also do another video with this new power adapter.

one thing i have t6 trimmer turned fully clockwise for the most notable flange., this is a wet dry mix trim?? is it not? if so from reading different notes most have it trimmed in the centre. is how ive set my t6 up similar to anyone out there???
i did one a few days ago and after uploading it for a few hours on you tube, it then gets rejected becauseover a gigabyte!

heres another link of next video

http://www.youtube.com/user/njkmonty1#p/a/u/0/sYSccDFzIO4

for some reason came out pixelly?

after that one is another testing sounds and trim pots

njkmonty · « **Reply #306 on:** July 13, 2011, 08:38:31 AM »

http://www.youtube.com/user/njkmonty1#p/a/u/0/--xoMQ9D-Oc

12Bass · « **Reply #307 on:** July 13, 2011, 09:18:43 PM »

Congratulations njkmonty, from your videos it would appear that you've got things mostly working! One thing to remember about this design is that it provides a very large range of adjustment and that achieving the best sounds requires careful adjustment of the controls (and that's after everything has been calibrated). I seldom use full LFO depth, as that sweeps too broad a range unless using extremely slow sweep rates.

The Enhance control adds regeneration (feedback) and its effect should be very noticeable - the sound should become very resonant and metallic as more Enhance is added. Maybe check T2 and the wiring to the Enhance pot and make sure the connections are functioning. I've set T2 so that maximum regeneration is just past the oscillation point (set to taste).

To set T1 (BBD bias), I disconnected the dry path and listened to the BBD path with a constant tone (generated by PC signal generator) and set a static clock (no LFO) then adjusted for highest output with least distortion. A sine test tone in the range of 60 - 500 Hz should suffice. Just listen for the onset of clipping and make increasingly slight adjustments until distortion is minimized. Optimal BBD bias varies somewhat according to clock frequency, so I tried to strike a balance somewhere in the middle range. Having a constant regulated board voltage is important to keep the BBD within spec.

T4 sets the base BBD clock, while T5 sets the range from low to high. These two are interactive. On my build, I set T4 to achieve roughly 13 ms at maximum delay (LFO depth at minimum). If regeneration is added, a 14 ms delay produces a resonant pitch which is around C#2 on a piano and sounds somewhat like a Cylon from Battlestar Galactica. T5 can then be used to set the minimum delay, which should sweep up to around 2 KHz at the highest (around C7). Using various tools and techniques, these controls can be adjusted without need of a frequency meter.

T3 adjusts the "lumpiness" of the LFO sweep. One trick to set the LFO for a fast rate, then adjust T3 to where the pulses are smoothest. This should give a reasonably smooth sweep when LFO speed is lowered.

As for T6, this didn't exist on my build (SAD1024A), though I added a small trimpot on mine for fine tuning. What I did was feed the flanger a pink noise signal from my PC, observed the output using an RTA program, and set the trimmer for maximum cancellation (largest visible nulls in frequency response) - you can see an illustration of this in an earlier thread where I posted the results. Again, this will vary somewhat according to clock rate due to non-linearities with the BBD, so I set mine for maximum cancellation through the "sweet" part of the sweep (somewhere between 0.5 ms and 2 ms).

12Bass · « **Reply #308 on:** July 13, 2011, 10:13:30 PM »

Because frequency and period are reciprocal relations, it can be calculated that 0.5 ms delay equates to roughly C7, 0.4 ms is roughly D#7, while 0.3 ms is roughly G#7. So, if regeneration is added, it should be possible to set the maximum clock rate to generate these pitches rather than using a frequency meter. In terms of clock frequency, 0.5 ms is around 1 MHz [1/(0.5 ms/1000/512 samples)], 0.4 ms is around 1.3 MHz, and 0.3 ms is around 1.7 MHz. My SAD1024 version is set for a minimum delay of roughly 0.2 ms, but this clock rate (2.6 MHz) may be a bit much for the MN3007.

njkmonty · « **Reply #309 on:** July 13, 2011, 11:42:16 PM »

heres another vid

http://www.youtube.com/user/njkmonty1#p/a/u/0/mO0BC6C9OBw

wow thanks for the answer!

its helped already and ive onlyabsorbed 20% of it!!, will read over again after printing it out, so what i gather is im close and just need to fine tune it better,
i was a little unsure if i was suppose to get extreme flang out of it.
i do get some pleasant chorus though, that led me to believe id stuffed up!

12Bass · « **Reply #310 on:** July 14, 2011, 01:31:32 AM »

Quote from: njkmonty on July 13, 2011, 11:42:16 PM

... so what i gather is im close and just need to fine tune it better,
i was a little unsure if i was suppose to get extreme flang out of it.
i do get some pleasant chorus though, that led me to believe id stuffed up!

Chorus is sort of a "sub-species" of flanging which occurs at longer delay times where you hear pitch modulation more than comb filtering. Aspects which make flanging more "extreme" are a large sweep ratio, very short minimum delay, longish sweep period, and high levels of regeneration. When properly set up, the A/DA hits all of these points. Also important is setting the wet/dry ratio to as close to 1:1 as possible, to maximize comb filtering. The best way to hear the impact of comb filtering is to feed the flanger a full range signal, or something with a lot of harmonics (like raw fuzztone guitar).

The range control adjusts the LFO depth, or how much the LFO modulates the delay time. At minimum, the LFO has no impact on the delay, and there's just a static delay, or a "frozen" flange sound. At maximum, the LFO sweeps the flanger from maximum to minimum delay time (as set by T4/T5).

Manual sets the starting point for the sweep. In other words, it sets the baseline delay from which the LFO moves up and down (mostly down, IIRC). So, if you set the manual control for 5 ms, the flanger will sweep up and down around that point, with the depth/intensity of the sweep controlled by the range knob.

Enhance controls regeneration or feedback. It's what makes for peaky/metallic/extreme oscillation effects. If there's a problem with the wiring here, it might explain why your flanger is not sounding as extreme as it should. By adjusting this control and T2 to maximum, you should be able to produce massive self-oscillation.

This thread contains a few samples from my SAD1024A A/DA build (note: "white noise negative" seems to have been deleted): http://www.diystompboxes.com/smfforum/index.php?topic=83352.0

12Bass · « **Reply #311 on:** July 14, 2011, 01:44:05 AM »

BTW, have you installed SW1? I'm hearing negative flanging in your videos, but not positive. Negative flanging sounds sort of "tubular" but won't produce the same "jet swoosh" as positive flanging. If not, try installing SW1, or just put a jumper across the pads and see what happens.

For my taste in flanging sounds, I tend to keep the delay out of the chorus range, sweeping from around 0.3 ms to say 8 ms. If you can, try feeding the flanger pink or white noise, or a full-range music signal, and play with the manual and range controls to fine tune the sweep. This should help familiarize yourself with the controls and is easier than tweaking while playing guitar. Having too much range will put the bottom end of the sweep into the chorus zone and sound "warbly" rather than "flangey". Also, see my earlier post regarding making the LFO less "lumpy".

njkmonty · « **Reply #312 on:** July 14, 2011, 06:54:57 AM »

one quick thing i noticed capacitor 18 asks for 1500pf?

i have a 1noj 100v cap in there. (actually i think im suppose to have 0.0015uF is that right?

does 0.0015uF = 1500pf?

also the only other value that was away from exact pdf specs was i used a 560pf cap instead of 510pf in c19?

you are right, im not getting much response from enhance pot

12Bass · « **Reply #313 on:** July 14, 2011, 07:06:12 AM »

Yes, 1500 pF = 0.0015 μF. The slightly larger 560 pF cap for C19 will make the regeneration path a little darker, while 1 nF C18 will make it a little brighter. FWIW, in the thread linked above, I noted a number of changes I made in my build to open up the sound by relaxing the low pass filtering. But that's stuff to think about after you've got a fully functioning flanger....

I'd suggest troubleshooting the enhance control. There may be a bad connection somewhere. Its effect should be obvious. No enhance implies that there's a break in the feedback loop somewhere (the circuitry surrounding IC1d and going to the enhance pot. Check continuity along the way with a multimeter.

sleepy · « **Reply #314 on:** July 21, 2011, 02:35:06 AM »

My ADA flanger is almost there.
I feel like it's pretty close just calibrated by ear, but I would like to check my work.

So, I've got a question regarding the calibration procedure for min and max frequency of the clock circuit.

I have an oscilloscope and a tone generator, but NOT a frequency counter.
However, If I send a tone to the scope which is roughly 69.6 KHz i can easily locate the correct time window.
So it would seem that I should be able to tweak the trim pots until my clock squares up in the window...

Doing it this way should work, but it would help to know roughly the voltage I'm looking for.
If I calibrate my window for 1v p-p I'm not really catching the clock signal...
Or if I am I'm not sure what I'm looking at, for.
Does it ride up on top of the 12.3 - 9.2volts at pin 13 of the 4047 or should I look for something else?
Square wave? sin wave? Approximate voltage p-p? 1volt? 1mv? .01mv?
I'll have another look tomorrow evening, but I could sure use a little bit of guidance.

Thanks a ton.
Sleeper

4007 4047 3007 4049
1- 0.34 1- 12.2 - 9.1 1- 13.93 Pins 1 & 16- 14.25
2- 0.34 2- 0.7 - 3.9 2- ~6.3 Pin 8- 0
3- 0 3- 7.4 3- 7.99 Pin 13- NC
4- 0.04 4- 14.25 4- 0.87 ALL other pins- ~6.37
5- 0.04 5- 14.25 5- 0
6- 0 6- 14.25 6- ~6.26
7- 0 7- 0 7- 4.9 - 6.0
8- 0.05 8- 0 8- 4.9- 6.0
9- 0.37 9- 0
10- 5.1 - 3.7 10- 5.9- 7.2
11- 7.2 - 7.8 11- 5.9-7.2
12- 0.7 - 3.2 12- 0
13- 14.25 13- 12.3 - 9.2
14- 14.25 14- 14.1-14.5

12Bass · « **Reply #315 on:** July 21, 2011, 04:10:32 AM »

IIRC, the clock should be a square wave which is 15V p-p at pins 2 and 6 of the MN3007.

oldschoolanalog · « **Reply #316 on:** July 21, 2011, 04:18:54 AM »

Yes. 15V p-p. CMOS gives you the rails out. Also, you can use a 'scope and a bit of math to do f measurements. Not super exact, but certainly more than accurate enough for this project.

sleepy · « **Reply #317 on:** July 21, 2011, 02:32:21 PM »

Thanks OSA and 12Bass
I was a bit under on my voltage range...

Tis no wonder that I missed seeing it.
I'll have another go this evening.
Thanks!

njkmonty · « **Reply #318 on:** September 14, 2011, 04:24:41 AM »

hello again i managed to get the enhance pot working (one pad not soldered!) woohoo!
will maybe do another YouTube. got the calibration pretty easy, now its working prop, using freq thingy on dmm

just looking at doing some mods (only usable ones0, could someone perhaps suggest whats worth doing and not?

i was looking at the stereo mod, but came across some "bounce" mod etc.
anyone had experience with any additions?

Fender3D · « **Reply #319 on:** September 14, 2011, 04:52:49 AM »

If you have not a specific objective to reach, even yellow knobs are an interesting mod

News:

Author Topic: MN3007 ADA Flanger Clone Questions (Read 224771 times)