Joined: Jun 29, 2007 Posts: 1113 Location: Loss Angeles
Audio files: 3
Posted: Thu Jun 05, 2008 5:00 pm Post subject:
TH VCO-1 Mods: Functional Subject description: Modification to use THAT340 PNP/NPN set
Find attached a redraw of the TH VCO-1 PCB artwork to use the PNPs in the THAT340 for the expo pair, and the NPNs for the sine shaper. Note that the 3386-type trimpots have been redrawn to utilize the inexpensive 100r, 47K, and mini 100K trimpots at All Electronics. Which is hilarious as I've replaced 10 cents in transistors with a $4.90 IC...
I've left space to use a 2K Tempco, and to place it across the PNP area of the THAT IC if required. I'm not using a Tempco in mine and don't hear Moog-style drift.
I've made space for a saw shaper but I first wish to confirm the expo FM character of the device prior to going any further. Thanks to Wayne Kirkwood/KA-Electronics.com for the THATs. Thank you again Thomas and Scott.
Note that the traces are a tad fat and bridged in a few places when etched. Thanks to Bubblechamber for showing me how, and also for the hint that the EN129 VCO shows a 3080-based sine shaper and triangle to saw shaper.
As ryktnk said in this thread, the THAT NPNs wouldn't/didn't really clean up the sine a great deal. The sine shape might be useful to bring out to a panel control, however.
Also, the Expo FM has an odd artefact- at certain pitches, and pitch sweeps, you can hear the modulator tone clearly, instead of harmonics. And as you'd expect from using a very fast, highly-matched PNP set for the expo converter, as well as 13V/uS slew on the 072s, the harmonics are very bright. Not what I'm after. You may love it. Losing the 100K resistor in the Expo input path really opens up the modulation range, as you might expect.
Next up will be using a 1458 in the core and sine shaper. Let's see what 0.5V/uS slew will do...(the Buchla uses that on the audio output; this will likely screw up the TH core!)
Note that I'm using a CA3080E from EEEDESIGNER on eBay, one of the $1.29 batch I'd recently posted about. I believe that they still have some (14) left. $3.00 flat-rate shipping around the world!
I see that it's been downloaded a few times since I've confirmed functionality... Please note that the jumpers should be installed as indicated: The blue jumper should be last, and should be installed after the
resistors are all stuffed, or at least the 2.2K it flies across.
6/13: A 1458 in the core works just fine, and does of course darken the sine wave (evident with deep index exponential FM) and the square wave slightly as well (not as much as a 1458 in the square section, but it might be just right only in the core). Losing the 100K input limiting expo resistor opens the door to a desired level of FM. It also allows the oscillator to go to ground when the pot is at "minimum" Adding a .47 film cap ala' the Buchla 258a FM input helps give a slightly better FM character as well as solving the ground problem.
The PWM range is a bit low. I'm going to swap out resistors and see...
Why not solder a jumper to the legs of the THAT340 across the top, to connect
the emitters ?
This too is ugly, but no need for extra tracks.
On your PCB layaout you appear to have connected the
emitter of the NPN to the PNP on the opposite side, this
should be connected to the NPN on the same side of the chip
Also for the sine shaper it does'nt seem to matter to much about the quality
of the PNP pair for this, and their is a trimmer for adjusting the
symmetry between the two transistors.
You can get a fantastically good sine wave with "vanilla" transistors.
But your idea of using the THAT340 is clever as it has the NPN & PNP pair.
I have included a JPG showing the harmonic content of the sine
output using just regular PNP transistors.
PS of the HF trimmer feed into the OPAMP, using THAT320, I found
that I had to lower the value of the 2m2 resistor to 1m0, although
I built it with LM13700.
-ryk
THVCO1_SINE_TRANSISTOR.jpg
Description:
Filesize:
85.9 KB
Viewed:
907 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
Joined: Jun 29, 2007 Posts: 1113 Location: Loss Angeles
Audio files: 3
Posted: Sun Jun 08, 2008 12:22 pm Post subject:
ryktnk wrote:
Hello
Hey.
Quote:
Why not solder a jumper to the legs of the THAT340 across the top, to connect
the emitters ?
This too is ugly, but no need for extra tracks.
I can see where that could happen, thanks. I choose to keep it underneath the IC as I always use sockets, and they have adequate clearance.
Quote:
On your PCB layaout you appear to have connected the
emitter of the NPN to the PNP on the opposite side, this
should be connected to the NPN on the same side of the chip
Also for the sine shaper it does'nt seem to matter to much about the quality
of the PNP pair for this, and their is a trimmer for adjusting the
symmetry between the two transistors.
You can get a fantastically good sine wave with "vanilla" transistors.
But your idea of using the THAT340 is clever as it has the NPN & PNP pair.
I'm sad to learn this- there is too much distortion/harmonic content in the sine wave for my tastes (which are not anal- I'm not expecting the characteristic of a self-oscillating filter from any oscillator).
I was hoping to save money in not using an SSM for the expo pair plus vanilla sine-shaper trannies. A THAT340 is less than just an SSM.
Quote:
I have included a JPG showing the harmonic content of the sine
output using just regular PNP transistors.
Joined: Jun 29, 2007 Posts: 1113 Location: Loss Angeles
Audio files: 3
Posted: Thu Jun 12, 2008 8:17 pm Post subject:
Just a quick kick because it works. I didn't check out the linear FM or tri output, however. Pulse and PWM work just fine, but with a slightly lower PWM range than I'm used to (edit: because I'm not using 10V p/p modulators...). Great tone...even with a 1458 in the core (also darkens the sine for nice expo FM). Nice range on the Coarse control!
Thank you Thomas and Scott!
Added a sample of the VCO modulating the Technosaurus filter resonance (res pot at zero, Fc swept by envelope). TH sine at a nearly inaudibly high frequency. I don't think that the Selector filter blocks negative voltage in the resonance modulation; it's fun to use an inverter envelope to modulate this parameter as well...
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