Tremolo-matic X from Stompboxology

[soggybag]

I made a pcb for this a long time ago by hand. The pots and LED were not mounted to the board. Worked well and sounded great but, was a total rats nest!

This is a harmonic trem. It splits the signal into a high pass and low pass, and cross fades from one band to the other. The sound is a little like a tremolo and a little like a univibe.








I made a new board with the pots and LED mounted to the board along with strategically positioning the pads for off board connections. Huge improvement!

I did these boards at PCBWay, I noticed there is a Jump in price if your boards are over 100mm. My first revision of this board was slightly over. I made a couple changes to squeeze everything down to the 100mm limit and the boards were $5 for 10!

[soggybag]

This project uses a the NE571 dual compander as a dual VCA. This was out of production and was harder to get but, now it:s back in production by Cool Audio.

You can get this from ElectricDruid.net or StompBoxParts.com for about $2 these days.

I haven't tried the new chip. I have some of the old chips. I'm guessing it should be a drop in replacement.

[ElectricDruid]

Very nice. The new one is certainly a lot tidier.

Do the highpass and lowpass controls vary the sound much? They're only 1st order filters, so they're fairly subtle, but with something like this, the phase shifts are almost as important as the filtering action.

The 571 rather than 13700 is an interesting choice in the Tremolo X design too.

[soggybag]

The sound is pretty noticeable but not extreme.

It was fairly subtle until I figured out I had used a 100n in the high pass filter instead of 2n2. When I fixed that error it really woke up.

It's in my head to try this with a 13700. The 571 is limited 5v headroom if I understand it. Seems like the 13700 should almost drop right into this circuit, the chip is even layer out similarly to the 571.

[ElectricDruid]

The 571 is limited 5v headroom if I understand it.

Seems unlikely. Where did you see that? The Coolaudio V571 version datasheet says 18V down to 6V supply, which doesn't sound like 5V headroom, except perhaps at the lowest possible supply. I suppose it's possible (well, it *is* possible) that's it's only 5V headroom at 9V, but that's a bit weird.

[soggybag]

I could wrong about the 5v. It's just something I remember reading. This circuit uses a 5v reg. On closer inspection the 5v regulator seems to there to add a bias at pin 3. Which doesn't make sense to me since there isn't a bias on pin 14?

[PRR]

...The 571 is limited 5v headroom if I understand it...

Yes. The supply is 15V for testing (18V max but why push it?). With perfect guts, 15V p-p is 5.3V RMS.


The 3080/13700 can't be drop-in, because they have naked diff-BJT inputs and distort bad by 20mV. (The 571 works different and has a 20k resistor inside.)

[Processaurus]

Nice work! The prices at PCBway are way too cheap. Seems like they're in a price war trying to undercut their competition? 

How do your high pass and low pass controls work? How does it change the sound?

[soggybag]

I built this from the Stompboxology article. Here's the schematic. The low pass filter is built around C4 and the High pass is C9.



I've been using PCBWay for everything I build. The prices are good and they allow smaller trace sizes, colored, solder mask, and the boards are pre-tinned. Shipping is the biggest part the cost. To counter this I have been ordering more than one board so it splits the shipping costs.

[antonis]

I built this from the Stompboxology article. Here's the schematic.

Hopping for C14 deleted..

[soggybag]

I'm an amateur, but I recall there was some discussion about having a cap to ground on the output of an op-amp was bad practice. Looks like I can remove one component!

[antonis]

I recall there was some discussion about having a cap to ground on the output of an op-amp was bad practice.

https://www.analog.com/en/analog-dialogue/articles/ask-the-applications-engineer-25.html

Another way to see it as follows:
10μF exhibits 132R impedance (for 60Hz mains, hence 120Hz ripple) or 159R (for 50Hz mains)..
The above resistance have to be driven by op-amp output..
The bigger the output ripple the bigger the op-amp output current..
(in case of well regulated voltage divider Vref, there's no need for reservoir capacitor..)

[Processaurus]

I built this from the Stompboxology article. Here's the schematic.

Hopping for C14 deleted..

Regarding opamp buffered Vref: It is much stiffer than a typical (10k x2, or 47k x2)voltage divider, great if you have many loads on vref, but there can be some hiss without the cap.  Yes the cap is a bad load for the opamp to drive directly. If you use the C14 cap, put a 100 ohm resistor (minimum 1/4 watt) between the opamp out and the capacitor. You get the noise free vref, but with that nice low impedance and current capacity you would get with a wasteful low resistance voltage divider and BFC. It will sink low frequency loading on vref with the opamp, and high frequency with the cap.

[antonis]

In case you use a cap and a resistor, they better be placed in parallel across feedback loop..
(same hiss prevention LPF but better result, IMHO..)

[Processaurus]

Re: opamp buffered vref, a subtle refinement is taking the feedback for the opamp' s inverting input from the capacitor, rather than the opamp out. Allows the opamp to better monitor and compensate for the voltage at the cap, which is what you care about more.

[antonis]

I should make output resistor 100Ω, just to have peace of mind..

[PRR]

...taking the feedback for the opamp' s inverting input from the capacitor, rather than the opamp out. .....

?? Seems to me that -reduces- stability by adding another pole (~~750Hz) in the open-loop response.

And for those values and the TL07x it may do very little because these opamps already have about 150-200 Ohm internal output impedance.

(TL062 very similar)

[m4268588]

[Processaurus]

Is the problem that if enough current gets drawn out of the capacitor suddenly, and the voltage at the inverting input sags below vref, the opamp will increase its voltage over vref and overshoot?

And for those values and the TL07x it may do very little because these opamps already have about 150-200 Ohm internal output impedance.

(TL062 very similar)

That's interesting, but wouldn't omitting the series resistor before the 22uF cap on the output take us back to the unsatisfactory circuit, where the inverting input is seeing the big capacitor voltage after a series resistance from the output (internal resistors)?

Is the best configuration series resistor to capacitor, but output wired directly back to the inverting input? I think the capacitor is necessary for good audio performance, I've experienced noise problems (hiss) with buffered vref w/o the cap on the output. To be fair I've only used it with TL07x opamp.

[soggybag]

I updated my design to fit portrait orientation. I've been using PCBWay.com. They have a deal on boards smaller than 100mm. The landscape board stretched this limit. Portrait fits easily.



Here is some more pictures. 

http://www.super-freq.com/tremolo-matic-x-build-3/