Author |
Message |
flab
Joined: Feb 13, 2012 Posts: 65 Location: Glasgow
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sun Nov 03, 2013 1:20 pm Post subject:
|
|
|
I think it should work to use a pot to pan between pulse out and saw out.
I have built this VCO.
One thing though - for this design to work properly, you should use 4069UB. If you want to try a 4049 (which I would not suggest for this circuit) then make sure the number is 4049UB and not 4049B. The 4069 and 4049 are not identical. The buffered version (4049B) uses a series of 3 amplifiers per gate and does not have the same "linear amplifier" characteristics that the UB version will have.
Aside from that, the 4069UB is designed with symmetrical transistors (one set of two transistors for each gate). This symmetry causes the gate to be able to source as much current as it sinks. The 4049, because it was designed to drive TTL, sinks far more current than it sources.
In general, I've found that the 4069UB works better as a linear amplifier than the 4049B or 4049UB. This circuit uses at least one of the gates as a linear amplifier.
This means that while the VCO may work with 4049 (that is, it will probably oscillate), it will likely not have a linear response to CV as it should. It may also not have a straight ramp for it's output. If you don't care about pitch accuracy or linearity of the waveform, then this won't matter.
And one final thing: When I built this, I had trouble with weird noise on the pulse output. In conversations with Rene, we concluded that specific pins needed to be used for specific functions. I had wired mine differently than Rene - once I used the same pins as Rene, my unit began to operate correctly. This may or may not be true of all manufacturers of this part. Just a heads up in case you see "grass" or "hash" in the pulse output signal. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
Danno Gee Ray
Joined: Sep 25, 2005 Posts: 1351 Location: Telford, PA USA
|
Posted: Sun Nov 03, 2013 1:45 pm Post subject:
|
|
|
How is one to know which pins Rene used? |
|
Back to top
|
|
|
flab
Joined: Feb 13, 2012 Posts: 65 Location: Glasgow
|
Posted: Sun Nov 03, 2013 1:56 pm Post subject:
|
|
|
JovianPyx you made it clear ,im going to use some 4069ub 's then ,about the pins uses we should look at his layout then |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
|
Back to top
|
|
|
flab
Joined: Feb 13, 2012 Posts: 65 Location: Glasgow
|
Posted: Sun Nov 03, 2013 5:19 pm Post subject:
|
|
|
now thats beautiful , thanks , ill give this a try next week - first i have to order some 4069s |
|
Back to top
|
|
|
rene_schmitz
Joined: Oct 27, 2013 Posts: 16 Location: germany
|
Posted: Sun Nov 03, 2013 6:45 pm Post subject:
|
|
|
Danno Gee Ray wrote: | How is one to know which pins Rene used? |
Here is a link to the original article:
http://www.schmitzbits.de/vco4069.html
There is an image of the way I had done the layout on pad per hole board.
The difference to JovianPyx is the top and bottom side of the chip is swapped.
To flab's question about the mixing of waveforms:
I'd recommend adding the sawtooth buffer amplifer as depicted on above page, then the crossfading should be more natural, since both amplitudes are (almost) the same. Otherwise you'd also have a change in volume as you fade.
You merely need to string a linear pot between the outputs (the buffered saw, and the pulse output, each going to the end stops of the pot) and probably buffer the signal at the wiper with an opamp follower.
Cheers, René _________________ My site: http://www.schmitzbits.de |
|
Back to top
|
|
|
flab
Joined: Feb 13, 2012 Posts: 65 Location: Glasgow
|
Posted: Mon Nov 04, 2013 3:36 pm Post subject:
|
|
|
thanks a lot Rene , what a honour !as i said i ll try your design next week probably -ill use the buffered mods as you suggest and i also use Motohiko's synch input . love your designs
cheers |
|
Back to top
|
|
|
Danno Gee Ray
Joined: Sep 25, 2005 Posts: 1351 Location: Telford, PA USA
|
Posted: Mon Nov 04, 2013 7:13 pm Post subject:
|
|
|
Thank you very much Rene. |
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
Posted: Mon Nov 04, 2013 11:09 pm Post subject:
|
|
|
Hi guys
This looks simple enough for me to build, thank you for posting this
If there is a sound clip I really would like to hear this vco.
Q...
The transistors needs to be matched?
How many octaves I can play with it?
cheers. _________________ http://www.myspace.com/mgmtrance |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Tue Nov 05, 2013 8:02 am Post subject:
|
|
|
I can't give advice on the matching because I replaced the expo converter with a linear one, but my build of the VCO gives me 9 octaves that are in tune. I've read that other builders have gotten more. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
Posted: Tue Nov 05, 2013 9:52 am Post subject:
|
|
|
thanks for replying.
few questions if i may..
1. so in the linear cv inputs version i dont need to match the 2 X 2n3904?
2. say i would like to do the linear cv mod without the tri sub mod, should it look like the next pic?
3. i also see that some resistors values are changed, say i will not use the LFO mod (pink area), should i change the resistors values or changing the resistors are not related to the LFO area?
4. is there any Tempco or NTC involved? i just dont see it if there is.
out of curiosity, if there is no Tempco or NTC from where or how the VCO stay stable?
5. what would be the cap value on the Raw FM input?
6 . the Raw FM input is the FM's summing in the circuit? if it is and i want to add more FM's should i add the unknown value cap for each and every cv input i add?
sorry for lots of questions, asking cause my lack of knowledge .
cheers
Isak E. _________________ http://www.myspace.com/mgmtrance |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Tue Nov 05, 2013 10:38 am Post subject:
|
|
|
isak wrote: | thanks for replying.
few questions if i may..
1. so in the linear cv inputs version i dont need to match the 2 X 2n3904? |
No matching required.
isak wrote: | 2. say i would like to do the linear cv mod without the tri sub mod, should it look like the next pic? |
Yes (see image above)
isak wrote: | 3. i also see that some resistors values are changed, say i will not use the LFO mod (pink area), should i change the resistors values or changing the resistors are not related to the LFO area? |
If I recall correctly, there were resistor changes because of some problem with the PWM circuit. This was a long time ago, I'm not sure if the values I show in my redraw are the ones that were finally settled on.
isak wrote: | 4. is there any Tempco or NTC involved? i just do see it if there is. out of curiosity, if there is no Tempco or NTC from where or how the VCO stay stable? |
No tempco of any kind is needed because this circuit does not use an exponential voltage to current converter. The linear voltage to current converter is largely immune to temperature changes. (I say largely because there is no such thing is complete thermal immunity. However, there is a point where it doesn't matter and in this circuit any thermal sensitivity is so small that it is negligable.)
isak wrote: | 5. what would be the cap value on the Raw FM input? |
I don't remember (and the equipment is not accessible at the moment). I'd experiment, start with .01 or .1 uF and work from there.
isak wrote: | 6 . the Raw FM input is the FM's summing in the circuit? if it is and i want to add more FM's should i add the unknown value cap for each and every cv input i add? |
To be honest, I don't remember testing this. I'd put a 10K in series with the cap too. Each FM input would be done that way, a resistor and a cap in series.
isak wrote: | sorry for lots of questions, asking cause my lack of knowledge .
cheers
Isak E. |
One other point - this circuit (as I've said before) is NOT EXPO. There are reasons why you would want expo. The main reason is for pitch modulation with something like an LFO. If you modulate an EXPO pitch CV with an LFO, the modulation effect will be audibly the same regardless of whether the note you play is a high pitch or a low pitch. You cannot expect the same effect of a LINEAR pitch CV. With linear, the effect of the modulating LFO will be greater at lower pitches than at higher pitches. This is the main drawback of linear pitch CV synths and is why the gross majority of analog synths are EXPO and not LINEAR. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
Posted: Tue Nov 05, 2013 12:32 pm Post subject:
|
|
|
Thank you for explaine this to me!
From the last part you wrote if I want to use LFO, EG and so, you "advice" me to stick with the original scheme? Correct?
Thank you again! _________________ http://www.myspace.com/mgmtrance |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Tue Nov 05, 2013 12:58 pm Post subject:
|
|
|
isak wrote: | Thank you for explaine this to me!
From the last part you wrote if I want to use LFO, EG and so, you "advice" me to stick with the original scheme? Correct?
Thank you again! |
Yes, that is correct.
I will assume that you already have a controller or sequencer that makes expo pitch CV.
While it is possible to make linear work like expo with regard to pitch modulation, it is rather complex. I worked out a design for this, but never actually built it. This is precisely why the PAiA Fatman synth has no LFO for pitch modulation. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
Posted: Tue Nov 05, 2013 1:18 pm Post subject:
|
|
|
Thank you my friend, it was very informative
I just checked Motohiko's site, he built this as well and used only 1 NTC for the cv inputs.
Have a look ...
http://www.aleph.co.jp/~takeda/radio/img/4069VCO-scheme.gif
he said in his site that Q1 and Q2 are matched, I think he meant Q1 and Q3 when looking at his scheme, don't you think?
I think they should be matched after all, haaa? _________________ http://www.myspace.com/mgmtrance |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Tue Nov 05, 2013 1:53 pm Post subject:
|
|
|
isak wrote: | Thank you my friend, it was very informative
I just checked Motohiko's site, he built this as well and used only 1 NTC for the cv inputs.
Have a look ...
http://www.aleph.co.jp/~takeda/radio/img/4069VCO-scheme.gif
he said in his site that Q1 and Q2 are matched, I think he meant Q1 and Q3 when looking at his scheme, don't you think?
I think they should be matched after all, haaa? |
Q1 and Q3 are involved in expo current generation and not Q2, so I would agree with the caveat that I am profoundly ignorant of good expo technique (because I don't use expo ) _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
|
Back to top
|
|
|
rene_schmitz
Joined: Oct 27, 2013 Posts: 16 Location: germany
|
Posted: Wed Nov 06, 2013 6:59 am Post subject:
|
|
|
You can exchange the 680 ohms resistor at the integrator for a say 1k trimpot and get a proper HF Trim adjustment. That might help to squeeze out the most out of the circuit.
I don't think that matching is affecting the tuning curve, it is R(bulk) that is affecting that. And these small signal transistors have pretty low R(bulk) values.
I think why Motohiko used a single NTC was that he couldn't get these easily, so I sent him some in exchange for tempcos. But the temperature compensation will be more accurate when you use the three NTCs.
@JovianPyx: Could the reason for the different resistors in the PWM section be that the FatMan uses different voltages for the negative and positive rail?
Cheers,
René _________________ My site: http://www.schmitzbits.de |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Wed Nov 06, 2013 7:27 am Post subject:
|
|
|
rene_schmitz wrote: | @JovianPyx: Could the reason for the different resistors in the PWM section be that the FatMan uses different voltages for the negative and positive rail?
Cheers,
René |
Quite probably so, the Fatman uses +8 and -12. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
Posted: Wed Nov 06, 2013 8:19 am Post subject:
|
|
|
Thanks Rene.
Now I know for sure not to match transistors
I'll replace the 680ohm to 1k trim like you advice.
Cheers. _________________ http://www.myspace.com/mgmtrance |
|
Back to top
|
|
|
flab
Joined: Feb 13, 2012 Posts: 65 Location: Glasgow
|
Posted: Wed Nov 06, 2013 10:41 am Post subject:
|
|
|
ooo ,the conversation went a bit further ,
Isak are you following Motohiko's inputs? -
ithink i ll have a go to design a wee pcb layout next week ,ill include that 1k trim pot as well |
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
Posted: Thu Nov 07, 2013 12:51 am Post subject:
|
|
|
Hi flab.
I'm going to use Rene's original scheme.
When looking at the Japanese site I didn't saw the sync that you talked about.
Can you show please? _________________ http://www.myspace.com/mgmtrance |
|
Back to top
|
|
|
rene_schmitz
Joined: Oct 27, 2013 Posts: 16 Location: germany
|
Posted: Thu Nov 07, 2013 5:06 am Post subject:
|
|
|
The sync input would be C4, R27, R12 in Motohikos scheme. Near the two schmitt trigger gates.
Cheers,
René _________________ My site: http://www.schmitzbits.de |
|
Back to top
|
|
|
isak
Joined: Dec 13, 2009 Posts: 847 Location: Israel
Audio files: 18
|
Posted: Thu Nov 07, 2013 5:38 am Post subject:
|
|
|
Thanks Rene.
I have additional questions please.
1. why R27 is a Pot? is it for the amount of the sync?
2. Can I use the sync parts as is in the original scheme?
3. Connected To pin 3 of the 4069, right?
4. I will want 2 or even 3 of this vco, if I want to sync between them I will need a sync out, is it possible to do, if yes how?
5. What would be the Vpp output?
As you can see I'm thrilled of making this vco
cheers,
Isak E. _________________ http://www.myspace.com/mgmtrance |
|
Back to top
|
|
|
|