simple saw to triangle shaper

[numbertalk]

In another thread someone was looking for a triangle-to-saw converter and someone mistakenly posted a saw-to-triangle circuit, which was actually a snippet from a larger circuit (the Obesifier - http://home.comcast.net/~sbernardi/elec/og2/og3_obesifier.html - see page 3). What I'm wondering is, if I just wanted to use the saw-to-tri converter circuit for my own purposes, looking at the diagram I'm attaching, I could eliminate R36A & R8, keep the rest of this snippet of circuit and get a triangle wave from the output of the TL71, right?

Another question is I see the Obesifier expects a 10Vp-p saw input - looking at the component values here would this sub-circuit work just as well with smaller saw waves?

---

circuit.jpg
Description:
Filesize:	7.16 KB
Viewed:	885 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[Dave Kendall]

Hi,

Well, here's a funny thing - I'm looking for the same thing, to drive saw animators in some experimenting........ small world isn't it?

I found this this while googling away. It's dead simple and low parts count.
I might only need one OPA, i.e. the first + second stages, as I want to turn a 0V to+7V (ish) saw into a 0V to +5V tri.

Haven't tried it yet, but I might get around to breadboarding it tonight - fingers crossed.......

cheers,
Dave

[richardc64]

[yusynth]

richardc64: You'd gain in replacing the 1N4148 by a Schottky diode with a low direct voltage drop.

[numbertalk]

Thanks for all the info. Some things to breadboard when I'm ready. Dave, how did your breadboarding go?

I actually kind of took this and ran with it from a posting ericcoleridge made on the thread about AndyR1960's ARP Odyssey VCO circuit, (http://electro-music.com/forum/topic-28604-125.html) so I'd be adding it to that, as I would also find a triangle wave useful in LFO mode. So not only will it not be a 10Vp-p wave but it also won't be centered around 0V. This is all fine - I can add DC offset and amplify the signal from the VCO output with an external module if necessary, I just hope I don't have to do that before feeding it to one of these wave shaper circuits. It would be nice to simply tap off of the saw outs and feed those into a couple these simple circuits and add a tri outputs to the panel from there.

[numbertalk]

Yves, do you have a part # you'd recommend for this diode?

Thanks!

[yusynth]

[numbertalk]

Thanks!

quote="yusson"]

[Dave Kendall]

Hi all.

I finally got around to wiring up the circuit below. It works well, but the input saw needs to be *exactly* centred around 0V for it to work glitch-free. The saw I was working with is a bit odd (+14V to +5V !), so I used an offset/scaler to centre it and bring it up to±5V.

The glitch filter part of the circuit removes a tiny spike at the top of the triangle. The output buffer is essential, as any load distorts the waveform. It can still be trimmed out, but a buffer is a good idea anyway.

Even with careful trimming of the offset, there is still a slight "flat top" on the triangle, which I couldn't trim out.

(PS my camera is up the swanee, so no scope traces for now....)

cheers,
Dave
(edited to correct info on the glitch filter)

---

sawtotri.gif
Description:
Filesize:	4.09 KB
Viewed:	799 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[Dave Kendall]

Hi All.

I tested both circuits posted in the thread - they are both extremely sensitive to offsets, and both require a saw centred around 0V as the input. I added an inverting offset trimmer to deal with this.
I couldn't trim out the "flat top" from the first circuit, and the second one has a nicer triangle, and only uses 2 OP amp stages instead of 3 (with the added offset) so I tested an adapted circuit based on the second one.
Originally its output was fully negative going only, so again an offset was added.

Advantages: it will take a positive-going saw, or one centred around 0V, (and probably a negative-only saw) it has variable gain, and will output a Triangle anywhere from negative going only, to centred, to positive going, depending on the setting of T4.
Disadvantages: The trimmers pretty much all interact, and to set it up, you really need an oscilloscope....

Put the scope on the output of the left OP amp, and set T1 so that the saw is centred around 0V. Then set T2 to give the desired gain - say, a 10V swing. (note; after testing with an EFM VCO4E saw (0V to +7.5V), T2 in the prototype was replaced with a 150K resistor.)
Then put the scope on the main output, and adjust T3 to get a nice triangle. If the triangle is lopsided, with one ramp longer than the other, adjust T1 and then T3 again until it's symmetrical. The setting of T1 is crucial to getting rid of a glitch, that shows up as a "step" or "spike" at the bottom of the triangle.
T4 has a very small effect on the triangle symmetry/glitch, so set that to get the desired offset (centred or positive), and then finally re-trim T1/T3 if necessary.
To cut down on the number of trimmers, you could install a temporary pot or trimmer in place of T4, trim it for the desired offset, and then replace with a pair of fixed resistors. T1/T3 will probably need a tiny final tweak.

This was tested by component substitution on perfboard, using ±15V, and seemed stable after leaving it on for several hours. Opamps used were TL072.

If anyone has spotted anything I missed, or has any suggestions, please post!

Enjoy, and Happy New Year to you all.

cheers,
Dave
(edited for typos)

---

NEW_SAW_to_TRI.jpg
Description:
Filesize:	230.61 KB
Viewed:	505 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[ericcoleridge]

I don't think this one has been posted yet, so here goes:

CGS VCO (with Voltage Controlled Saw to Tri + Sine)

This circuit, from the CGS VCO (lower left corner of schem) has voltage control over shape/symetry and uses just one op-amp (plus 1 op-amp for Sine conversion). It works independent of the VCO core; I've tested it on it's own, and it works great.

[Dave Kendall]

[ericcoleridge]

[ericcoleridge]

[numbertalk]

I emailed with Ken Stone about this part of his VCO circuit and he suggested using a precision rectifier circuit for this and approved of this one:

http://hem.bredband.net/bersyn/VCO/Saw%20to%20tri-sin%20W.gif

For those of us who lack protoboarding skills there's even a PCB image:

http://hem.bredband.net/bersyn/VCO/Saw2tri-sine_L.pdf

And component placement:

http://hem.bredband.net/bersyn/VCO/Saw2tri-sine_K.pdf

It's a quad board but you could obviously trim your P-n-P and etch a single circuit.

Here's further description of this circuit from his web site:

I developed this circuit, with the Moog 921B VCO as inspiration. It uses the classic one-transistor approach for folding the sawtooth wave into a triangle wave. But I added a special RC-network to cancel the glitch that comes from the less-than-instant reset time of the saw wave. I also replaced the emitter follower transistor with an op-amp buffer. This was necessary to get the output voltage up to 10 volts peak to peak. In addition, the circuit was adapted for +-15 volts supply, instead of Moog's +12 and -6.
The triangle to sine converter consists of two transistor pairs from a CA3046 array. Properly adjusted, this gives a very good sine wave. It sounds much cleaner than the diode-limiters that many synths use to approximate a sine wave. Again the emitter follower is replaced by an op-amp.

[JovianPyx]

http://home1.gte.net/res0658s/fatman/saw_to_tri_converter.html

[fonik]

...necroposting... (i love this word)

actually not that simple. anyways, yesterday i built this in spice during my lunch brake after reading jim patchell's ideal diode circuit. it will work with falling/rising ramp, even with only going positive ones (that's what i need it for). however, i've had no chance to listen to it right now.