| Sell a simple, cost-effective, stable, accurate VCO? |
| Where have you been all my life -- this is exactly what I have been waiting for! |
|
77% |
[ 21 ] |
| No, I have too many VCOs already. |
|
0% |
[ 0 ] |
| No, I only like VCOs that drift and don't track over more than 3 octaves. |
|
3% |
[ 1 ] |
| No, your external waveshapers are too weird. |
|
7% |
[ 2 ] |
| No, we don't use VCOs around here. |
|
7% |
[ 2 ] |
| What's a VCO, and why would I want one? |
|
3% |
[ 1 ] |
|
| Total Votes : 27 |
|
| Author |
Message |
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
 Posted: Tue Jan 27, 2009 6:24 pm Post subject:
Re: Sync? |
  |
|
| Guatis wrote: | | What would be the best possible way to implement an external sync input on you SAW CORE? . Should it be something similar to the MFOS Vco? Any thoughts on this? |
I won't be able to actually to try this, but you should be able to get sync by kicking A3 pin3,al la ENS-76 Option 1:
1.) Remove C10
2.) Increase R30/R31 to, say, 33k and 120k.
3.) Feed the sync signal directly to pin3 through a small cap, say 100p.
The "best possible way" would be to add an opamp comparator to the above, to get a consistent pulse into the cap.
Let us know if you come up with something specific that works for you.
MFOS? MFOS?!?!?!?! 
Ian |
|
|
Back to top
|
|
 |
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Tue Jan 27, 2009 6:28 pm Post subject:
|
|
|
| Fetafarmer wrote: |
Well, why should one person's statements be enough to kill off a project for a whole group of interested people? |
Because it killed off any significant interest? Your "whole group of interested people" was less than 2 dozen.
I went after management with an ax over this one, but got nowhere. If trolling bothers you, maybe you should also speak up.
Ian |
|
|
Back to top
|
|
|
guatis
Joined: Jan 18, 2009
Posts: 10
Location: Philippines
|
| Posted: Tue Jan 27, 2009 7:04 pm Post subject:
Re: Sync? |
|
|
| frijitz wrote: |
I won't be able to actually to try this, but you should be able to get sync by kicking A3 pin3,al la ENS-76 Option 1:
1.) Remove C10
2.) Increase R30/R31 to, say, 33k and 120k.
3.) Feed the sync signal directly to pin3 through a small cap, say 100p.
The "best possible way" would be to add an opamp comparator to the above, to get a consistent pulse into the cap.
Let us know if you come up with something specific that works for you.
MFOS? MFOS?!?!?!?!
Ian |
Thanks for the tips. I really don't care about extra parts like opamps. Just the best method. I'll try it
Cheers! |
|
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007
Posts: 1536
Location: nyc
Audio files: 41
|
|
|
Back to top
|
|
|
loss1234
Joined: Jul 24, 2007
Posts: 1536
Location: nyc
Audio files: 41
|
|
|
Back to top
|
|
|
guatis
Joined: Jan 18, 2009
Posts: 10
Location: Philippines
|
| Posted: Wed Jan 28, 2009 7:54 am Post subject:
|
|
|
| loss1234 wrote: | btw...sorry if this is an obvious thing but what are D2, and D3 (lm33365) doing along with c3 and c4..is it some sort of regulation or reference? i dont see it connecting to anything but the rails and gnd.
thanks |
They provide an stable 5.0 V and -5.0 V references for the critical parts in the circuit (voltage across frequency pots, comparator reference and current source of the expo converter). check for the up/down white arrows.... |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Wed Jan 28, 2009 8:48 am Post subject:
|
|
|
| Guatis wrote: | | They provide an stable 5.0 V and -5.0 V references for the critical parts in the circuit (voltage across frequency pots, comparator reference and current source of the expo converter). check for the up/down white arrows.... |
Exactly! And I see there is a typo -- it's LM336, not LM3336.
Ian |
|
|
Back to top
|
|
|
guatis
Joined: Jan 18, 2009
Posts: 10
Location: Philippines
|
| Posted: Wed Jan 28, 2009 8:55 am Post subject:
|
|
|
Ian,
I tested the circuit on breadboard last night and it worked after a few minutes of debugging. I noticed the R29 resistor gets quite hot during operation, and if i pull it out, the 15V rail current drops about 2mA (although the Core no loger works). Is this normal? I'm using +-15V supplies.
Tomorrow i'll be doing some measurements.
Best Regards |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Wed Jan 28, 2009 9:29 am Post subject:
|
|
|
| Guatis wrote: | | ...I noticed the R29 resistor gets quite hot during operation... |
Hey, you got it running already!
The circuit is for +/-12 V supplies. D5 reduces the LM311 supply voltage to around 6-7 V. For 15 V supplies you will need to change D5 to an 8.2V Zener.
Yes, R29 does get warm. I found I had to pull up the LM311 quite hard to get maximum switching speed. But you are putting about 1/4 watt through it, and with a lower supply voltage this is reduced to about 1/10 watt.
Ian |
|
|
Back to top
|
|
|
guatis
Joined: Jan 18, 2009
Posts: 10
Location: Philippines
|
| Posted: Wed Jan 28, 2009 10:25 am Post subject:
|
|
|
| frijitz wrote: | ...you will need to change D5 to an 8.2V Zener.
Ian |
Totally right. I should switch to a higher voltage zener to provide the 6-7V (and obviously adjust oher resistor values to get similar ranges)
Ben |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Wed Jan 28, 2009 11:20 am Post subject:
|
|
|
| Guatis wrote: | | (and obviously adjust oher resistor values to get similar ranges) |
I think everything else that is critical goes to +/-5V. Your results will be of interest to many folks here, so please keep us posted.
Ian |
|
|
Back to top
|
|
|
guatis
Joined: Jan 18, 2009
Posts: 10
Location: Philippines
|
| Posted: Wed Jan 28, 2009 1:06 pm Post subject:
|
|
|
| frijitz wrote: | | Guatis wrote: | | (and obviously adjust oher resistor values to get similar ranges) |
I think everything else that is critical goes to +/-5V. Your results will be of interest to many folks here, so please keep us posted.
Ian |
I'm planning to do a simple layout and mount the whole circuit on a self made PCB. Then i'll do several measures of frequency vs time, freq vs temperature perhaps, and volt/octave accuracy across the whole range. Then i'll plot the results and test the sync feature with a second VCO. |
|
|
Back to top
|
|
|
guatis
Joined: Jan 18, 2009
Posts: 10
Location: Philippines
|
| Posted: Wed Jan 28, 2009 6:58 pm Post subject:
|
|
|
Well. Today i did some stability measuring, and i got pretty good results with drift over time (~0.02%). Then i realized it doesn't make much sense to test drift over time, but instead drift over temp. So i did a rather extreme test using a hairdrier, switching between ambient temp and very hot settings (i didnt measure temp, but the components did get very hot). Obviusly the results were very diferent: up to 2% difference between the "cold" and "hot" frequencies @ 2kHz.
I know i should have prepared a more realistic setup, putting, for instance, the hole thing in an enclosure, to get more even temperatures across all the board. And of course, measuring the temp.
So, since i don't have space or components count restrictions, i think i'll give it a try to Ian's dial-a-tempco circuit.
So, Ian, how up to date is the info on your site (to schem is from 2003)? Have you done any improvements?
Any advice on how to use it with your present Saw Core implementation?
Many thanks for the help
BTW: I switched to a 8.2V zener with no problems. |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Wed Jan 28, 2009 9:07 pm Post subject:
|
|
|
| Guatis wrote: | | Then i realized it doesn't make much sense to test drift over time, but instead drift over temp. |
I wish all the people selling VCOs would recognize that.
| Quote: | | So i did a rather extreme test using a hairdrier... |
No, you cannot get reasonable stability, or any kind of reasonable drift measurement by using a hair dryer. At the level of stability achievable with a good design, you must absolutely make certain the board is heated uniformly and that there are no drafts.
Here's a fairly simple home method:
Put a semiconductor temp sensor on the board -- just kluge in the chip and a resistor and solder on a long output lead.
Put long leads on all the wires to the board, mounting all the controls at the far end.
Find a plastic kitchen container with a snap-on lid that is big enough for the board.
Make a sandwich in the container consisting of crumpled newspapers, the board, and crumpled newspapers. This will keep the board insulated from the outside enough that it will self-equilibriate to a reasonably uniform temperature, and it will prevent any large drafts.
Pull the leads out over the lip of the container and snap the lid on, right over the wires.
Now rig up a drugstore heating pad to warm the container. It works ok to make another sandwich with insulation (newspapers or towels), the pad, more insulation, the container, more insulation then towels draped across the top and down the sides. Warm this up very slowly (except maybe an initial cycle quickly up and down as a "bake-in"). I use a variac to set the power in the heating pad, but you can also just put the heat on for a short time and let it drift up to equilibrium. Be very patient doing all this. You won't get repeatable results unless you go real slow (hours).
I hope your pullup resistor is heating up less now. It can actually heat up the board and mess up your measurement. If your thermometer never comes back to room temperature, that will be why.
| Quote: | So, since i don't have space or components count restrictions, i think i'll give it a try to Ian's dial-a-tempco circuit.
So, Ian, how up to date is the info on your site (to schem is from 2003)? Have you done any improvements? |
I really suggest you try the above first and see if it isn't good enough. If your tempco resistor is right on you should have very little scale factor drift, which is the important effect.
The dial-a-tempco circuit on my website turned out to have a stupid restriction on frequency range. It's been redesigned, and I have two of them in my system right now, but I haven't written up the new circuit, which has different numbers from the original.
| Quote: | | Any advice on how to use it with your present Saw Core implementation? |
It shouldn't be any different from what I am using already. I would just need to dig out my notes and scribble up a drawing.
I'm glad to see someone interested in all this.
Ian |
|
|
Back to top
|
|
|
andrewF
Joined: Dec 29, 2006
Posts: 1176
Location: australia
Audio files: 4
|
| Posted: Wed Jan 28, 2009 10:06 pm Post subject:
|
|
|
Hi Ian
would it be possible to mod the cct to suit a 1k tempco? Just change R5 and R14 or is it a lot more involved than that?
cheers |
|
|
Back to top
|
|
|
guatis
Joined: Jan 18, 2009
Posts: 10
Location: Philippines
|
| Posted: Thu Jan 29, 2009 6:24 am Post subject:
|
|
|
 Wow. Surely i'll try this method with the Saw Core. I'll be more patient this time
A million thanks for your advice...
| Quote: | | The dial-a-tempco circuit on my website turned out to have a stupid restriction on frequency range. It's been redesigned, and I have two of them in my system right now, but I haven't written up the new circuit, which has different numbers from the original. |
Well, an update to your site (both Saw VCO and D-a-T pages) wouldn't do any harm i think 
| Quote: | Any advice on how to use it with your present Saw Core implementation?
It shouldn't be any different from what I am using already. I would just need to dig out my notes and scribble up a drawing.
|
If i'm having drift issues, that drawing would be a dream. But first i have some serious testing to do.
Ben |
|
|
Back to top
|
|
|
mono-poly
Joined: Jul 07, 2004
Posts: 937
Location: Rotterdam, Netherlands
Audio files: 2
|
| Posted: Fri Jan 30, 2009 4:10 am Post subject:
|
|
|
Ian you got 21 votes!
Aint that enough yet?
 |
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Fri Jan 30, 2009 10:39 am Post subject:
|
|
|
| andrewF wrote: | | would it be possible to mod the cct to suit a 1k tempco? Just change R5 and R14 or is it a lot more involved than that? |
You would need to also cut all the other resistors connected to the (-)input of U2a in half. Nothing wrong with that, but you no longer have the standard 100k input impedance for the module.
Ian |
|
|
Back to top
|
|
|
andrewF
Joined: Dec 29, 2006
Posts: 1176
Location: australia
Audio files: 4
|
| Posted: Fri Jan 30, 2009 5:50 pm Post subject:
|
|
|
Cheers, may get a bit messy
but I will give it a try. |
|
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005
Posts: 4127
Location: Mount Hope, KS USA
Audio files: 96
|
| Posted: Sat Jan 31, 2009 5:37 pm Post subject:
|
|
|
| Quote: | | 21 yeas looking for about 50 or so boards was more interest than the first Klee generated, and that's sold its 3rd round of 100 out since. It's too bad that a good project gets scuttled by Internet noise. |
I would have to agree with E-Rex here. Damn the torpedoes! Whites of the eyes! F the f'ing f'ers!
_________________
My Site |
|
|
Back to top
|
|
|
reve
Joined: Feb 23, 2008
Posts: 149
Location: USA
|
| Posted: Sun Mar 27, 2011 7:45 pm Post subject:
|
|
|
Couple questions here, folks.
(edit: I dropped the idea of doing a double board. Saved parts but didn't save a significant number of pins.)
Second: anyone get sync working well?
Thanks!
_________________
- r. mosquito
SynthDIY Synthpop -> http://www.reverbnation.com/photovoltaik |
|
|
Back to top
|
|
|
jordroid
Joined: Jan 17, 2010
Posts: 193
Location: ithaca, new york
|
| Posted: Mon Oct 24, 2011 3:27 pm Post subject:
|
|
|
It's been a while since you asked, but i just implemented sync as Ian recommended earlier in this thread and it works nicely, i couldn't detect much difference running the master VCO through a comparator or just straight into the cap, but ymmv. It's easy to try out anyways.
regards,
jordan |
|
|
Back to top
|
|
|
inlifeindeath
Joined: Apr 02, 2010
Posts: 316
Location: Albuquerque, NM
|
|
|
Back to top
|
|
|
inlifeindeath
Joined: Apr 02, 2010
Posts: 316
Location: Albuquerque, NM
|
|
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54
|
| Posted: Sun Feb 19, 2012 11:03 am Post subject:
|
|
|
The one on the website is earlier (1998 per the schematic). Both work really well.
Ian |
|
|
Back to top
|
|
|
|
Forum index » DIY Hardware and Software
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
