Help me add hardsync

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: lisa, luketeaford, Kent, Joe.

Post Reply
comrade_zero
Learning to Wiggle
Posts: 27
Joined: Tue Jan 22, 2013 4:15 am
Location: Arizona, USA

Help me add hardsync

Post by comrade_zero » Mon Jun 29, 2020 11:23 pm

So I have breadboarded David Haillant's simple VCO, (schematic linked below) and it works a treat! But, I would like to add hardsync if possible. I would prefer to avoid FETs (keeping it as parts sourcing friendly as possible), so if anyone has a good solution using a BJT or op-amp I would appreciate it. Thank you all in advance. Cheers!
http://www.davidhaillant.com/l/uploads/ ... 17b411.png

User avatar
guest
Super Deluxe Wiggler
Posts: 4215
Joined: Mon Aug 19, 2013 11:49 am

Re: Help me add hardsync

Post by guest » Tue Jun 30, 2020 1:48 pm

hard sync would go to pin5 on U1B. you want a short, upward pulse there for the sync. if your sync pulse is 10V, then just a capacitor and resistor will work. otherwise you will need another opamp or 2 transistors to condition the signal. the opamp will be more straightforward. you make a comparator out of the opamp with a threshold of 1V (or whatever you want it to be), and then put a resistor/capacitor/diode from the output to U1B.
openmusiclabs.com

comrade_zero
Learning to Wiggle
Posts: 27
Joined: Tue Jan 22, 2013 4:15 am
Location: Arizona, USA

Re: Help me add hardsync

Post by comrade_zero » Tue Jun 30, 2020 11:35 pm

Thanks, guest! Someone on the FB synthDIY suggested this conditioning circuit:Image and it works, but I do have a spare op-amp from a square/pwm waveshaper (not shown on the schematic above,) so getting an op-amp solution would actually be optimal. So if I have this straight a 100k/10k voltage divider on the negative input of the op-amp should set the threshold just over 1v, with the sync master signal into the positive input and the output going through an R/C highpass (cap, resistor to ground) and then through a diode into pin 5 of the circuit above? Please excuse my (extreme) lack of knowledge, but do I need a resistor before the positive input of the op-amp(?) and can you recommend reasonable values for the cap/resistor combo? My fooling around on the breadboard so far has not produced results other than heating up my TL074...
Thank you for your time and your help! Cheers!
You do not have the required permissions to view the files attached to this post.

User avatar
guest
Super Deluxe Wiggler
Posts: 4215
Joined: Mon Aug 19, 2013 11:49 am

Re: Help me add hardsync

Post by guest » Wed Jul 01, 2020 12:39 am

at the positive input, put a 100k to ground, so if nothing is plugged in, the circuit doesnt float.

for the highpass filter to the second opamp, you can use a circuit the same as the input to your transistor circuit above. i would probably drop the capacitor down to 47pF, though. and then you have the diode going from the 47k to the second opamp input. you dont strictly need a resistor in series, but a 1k or so wouldnt hurt, and would limit current into the input pins if the positive and negative inputs are at wildly different voltages.
openmusiclabs.com

comrade_zero
Learning to Wiggle
Posts: 27
Joined: Tue Jan 22, 2013 4:15 am
Location: Arizona, USA

Re: Help me add hardsync

Post by comrade_zero » Wed Jul 01, 2020 1:58 am

Thank you again for the help, guest! I *think* I got it sorted, but using the op-amp seems to drive the output amplitude of the oscillator pretty high (compared to the output without the sync). I don't have a 47pF on hand, so maybe I will come back to this when I do. For now I think I will stick with the transistor circuit above. I appreciate the help though. Cheers!
c_z

User avatar
EATyourGUITAR
has no life
Posts: 4774
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Help me add hardsync

Post by EATyourGUITAR » Wed Jul 01, 2020 9:08 am

Guest I was looking at this but I started thinking it should have some OR circuit. Some double latch. Some gate to trigger converter. How do we avoid making the VCO core stand still when sync has a gate high? Do we simply ac couple sync input? How do we control the cap charge phase so that it always ends when it would normally reset. The sync signal can happen at any time so a fixed length sync pulse seems more like FM. That's why all the latches and logic. What do you think?
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
guest
Super Deluxe Wiggler
Posts: 4215
Joined: Mon Aug 19, 2013 11:49 am

Re: Help me add hardsync

Post by guest » Wed Jul 01, 2020 9:50 am

all good points. the diode acts as the OR in this case, but really only works well if the pulse is very short (1us, w/ AC coupling), such that it is gone before the core can reset itself. the schmitt trigger comparator acts as the flip flop to save state for the circuit, so just a brief pulse to make it change state is required. this is the problem the OP is having, in that the pulse is too long, and its holding the flip flop state in reset too long, and the waveform is going above the top threshold.
openmusiclabs.com

User avatar
devinw1
Super Deluxe Wiggler
Posts: 1440
Joined: Tue Aug 07, 2018 11:20 am
Location: Portland, OR
Contact:

Re: Help me add hardsync

Post by devinw1 » Wed Jul 01, 2020 11:51 am

If you're gonna have 3 resitors, 1 cap, a diode, and those 2 transistors already, why not make a fixed-pulsewidth generating flip flop instead? Seems better for this.

User avatar
guest
Super Deluxe Wiggler
Posts: 4215
Joined: Mon Aug 19, 2013 11:49 am

Re: Help me add hardsync

Post by guest » Wed Jul 01, 2020 12:09 pm

you mean to replace the schmitt trigger? or to drive it? either way, not a bad way to go, but i think for this case, the OP has a spare opamp handy, and wants minimal changes (least time to synthing!).
openmusiclabs.com

User avatar
devinw1
Super Deluxe Wiggler
Posts: 1440
Joined: Tue Aug 07, 2018 11:20 am
Location: Portland, OR
Contact:

Re: Help me add hardsync

Post by devinw1 » Wed Jul 01, 2020 3:07 pm

I was thinking to drive it. Understood on the minimal changes part though!

KSS
Super Deluxe Wiggler
Posts: 2410
Joined: Mon Jan 25, 2016 7:28 am

Re: Help me add hardsync

Post by KSS » Wed Jul 01, 2020 10:07 pm

Hit quote instead of edit
Last edited by KSS on Wed Jul 01, 2020 10:16 pm, edited 1 time in total.

KSS
Super Deluxe Wiggler
Posts: 2410
Joined: Mon Jan 25, 2016 7:28 am

Re: Help me add hardsync

Post by KSS » Wed Jul 01, 2020 10:13 pm

The TH VCO1 you can read about here:
https://www.birthofasynth.com/Thomas_He ... VCO-1.html

has this to add to the discussion
"ThomasHenry on his VCO1 Design" wrote:IC3a also presents the triangle wave to the input of the Schmitt trigger; this is a discrete version built around transistors Q1 and Q2. I haven't taken this approach just to be a penny-pincher (and there's no doubt the transistor method is dirt cheap), but rather because I found I needed as much speed as possible. The various op-amps I tried as Schmitt triggers just didn't switch quickly or cleanly enough at higher frequencies. This isn't a trivial matter either, for as hinted at earlier, a lag in the switching time manifests itself in noticeable tuning imperfections and variability in the output amplitude of the triangle. In a nutshell, the discrete Schmitt trigger is about as good as they come, even if it does seem to be out of the stone age. Incidentally, I first saw this transistor application in "Improving the Linear VCO", Application Note Number 20, December 25, 1976, by the incredibly prolific Bernie Hutchins of Electronotes fame. It recently made a reappearance in the December 2003 issue of Nuts & Volts Magazine (p. 64). It's a technique that deserves to be better known.
That's only an excerpt and it's tri-core rather than saw, but the words hold up. Using an opamp as a comparator can work sometimes, but in a VCO little thngs add up. In either a good way or a not-so-good way.

Here's the related VCO core schematic from that link too. https://www.birthofasynth.com/Thomas_He ... schem1.pdf

-------------------
For a saw core, the discrete 4027-1 of the ARP 2600 is hard to beat for economy of design vs performance. Matched pair of the 3046-86 there is now easily replaced with inexpensive SMD. Hard sync of that core is well documented in the ODY schematics and TTSH add-ons.

User avatar
guest
Super Deluxe Wiggler
Posts: 4215
Joined: Mon Aug 19, 2013 11:49 am

Re: Help me add hardsync

Post by guest » Wed Jul 01, 2020 10:43 pm

thats a good comparator. when i was working on the high-frequency vco a while back, i needed a ridiculously fast switch of some sort, and looked at using this one. if i recall correctly, i found that the thresholds had some temperature drift due to mismatched base impedances (which is fixable) and mismtached collector currents at the match point (and associated Vbe differences). i wish i had written all this down, as i cant find it in my notes anywhere. i came up with a 3(?) transistor version that didnt have drift and was a hair faster, but that all got put to the side when i decided to use the 74HC221, which was very fast and accurate (on one of its thresholds, which was all i needed).

maybe it was a 5 transistor. a basic 4 transistor diff amp core with a PNP inverter at the output, collector outputting to one of the input bases for hysteresis. the currents are exactly balanced at the switch point. but the top switch point is a function of Vce,sat.
openmusiclabs.com

User avatar
EATyourGUITAR
has no life
Posts: 4774
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Help me add hardsync

Post by EATyourGUITAR » Thu Jul 02, 2020 7:32 am

Is slow CMOS related to the low power of CMOS? Is there no CMOS Schmitt trigger that is fast enough? I would imagine that the two limiting factors are internal charging current and pin or package capacitance.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
guest
Super Deluxe Wiggler
Posts: 4215
Joined: Mon Aug 19, 2013 11:49 am

Re: Help me add hardsync

Post by guest » Thu Jul 02, 2020 10:54 am

the input of the 74Hc221 is a cmos schmitt trigger, and was very fast. the issue i usually have with using them, is that the thresholds drift with temperature, and are different from part to part. if i recall, one of the thresholds didnt drift significantly, so it was fine for what i was doing. actually, the thing i was working on was a charge transfer core, so threshold level didnt matter much at all.
openmusiclabs.com

User avatar
devinw1
Super Deluxe Wiggler
Posts: 1440
Joined: Tue Aug 07, 2018 11:20 am
Location: Portland, OR
Contact:

Re: Help me add hardsync

Post by devinw1 » Thu Jul 02, 2020 11:12 am

I mean, the 74HC14 transition time and propagation delay at 5V Vcc is like 19ns...not terrible. Of course the faster CMOS stuff is usually 0-6V single supply which can be inconvenient at times compared to the older stuff which usually takes 20V.

KSS
Super Deluxe Wiggler
Posts: 2410
Joined: Mon Jan 25, 2016 7:28 am

Re: Help me add hardsync

Post by KSS » Thu Jul 02, 2020 7:04 pm

The other issue is using a big chip -most of it wasted- where something smaller will suffice. I understand the few components may well add up to similar footprint. But that discrete footprint can be massaged within the available PCB area -potentially resulting in better layout- where the chip is a block, period.

Post Reply

Return to “Music Tech DIY”