Author |
Message |
richardc64
Joined: Jun 01, 2006 Posts: 679 Location: NYC
Audio files: 26
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Tue Oct 10, 2006 3:52 pm Post subject:
|
|
|
Looks like it would work but it depends on the application. Give it a try. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
yusynth
Joined: Nov 24, 2005 Posts: 1314 Location: France
|
Posted: Wed Oct 11, 2006 2:53 am Post subject:
|
|
|
It should work. _________________ Yves |
|
Back to top
|
|
|
richardc64
Joined: Jun 01, 2006 Posts: 679 Location: NYC
Audio files: 26
|
Posted: Thu Oct 12, 2006 2:16 am Post subject:
|
|
|
Wow. 2 opinions not to be taken lightly. Thanks guys.
The application is to catch the right voltage at the right time from a DAC that's cycling through a keyboard scan counter. Adding in a 1V octave shift, or a pitch-bend, or tuning voltage -- which must be done at the DAC -- the final CV could exceed the analog switch supply voltages, which must be +/-5V. So I'm thinking scale the DAC output to 0.5V/Octave and bring it back to 1V/Octv after the switch.
I hope to post this idea circuit soon. |
|
Back to top
|
|
|
yusynth
Joined: Nov 24, 2005 Posts: 1314 Location: France
|
Posted: Thu Oct 12, 2006 2:29 am Post subject:
|
|
|
Well considering your application I would suggest that you separate the gain stage from the buffer stage, it costs another OPA to do so.
Cheers _________________ Yves |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Thu Oct 12, 2006 12:09 pm Post subject:
|
|
|
yusson wrote: | Well considering your application I would suggest that you separate the gain stage from the buffer stage, it costs another OPA to do so. |
I agree. This is one case where viewing an op amp as an ideal element can lead to subtile problems. In most cases it is safe to assume the op amp to have infinite input impedance and zero input offset voltage, but in any real device that isn't the case. The input resistance and/or offset voltage might change as you adjust the gain. Probably not a problem, but it might bite you.
BTW, the original Moog Modular keyboard had very poor S/H circuits. There were several variations over time, but on all of them that I tested there was a voltage drop or boost when the key was released. I mention this because a keyboard S/H is one case where it pays to get it right - don't economize. There are times where even the tinyist error shows up. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
richardc64
Joined: Jun 01, 2006 Posts: 679 Location: NYC
Audio files: 26
|
Posted: Fri Oct 13, 2006 1:55 am Post subject:
|
|
|
Dang it! The point was to keep the parts count down. There's going to be 2 of these S/Hs in the project. Back to the drawingboard.
There's already going to be another opamp(s) after the S/H(s) for portamento, so change the analog switch to a potentiometer and change the question to: Can a Lag have gain? |
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Fri Oct 13, 2006 9:25 am Post subject:
|
|
|
You can have gain in virtually any op amp circuit. Most of the time you can view the op amp as an ideal element, but when you are using it to measure a voltage on a capacitor, then the non-ideal effects might show their ugly head. I think it would be much safer to add gain to a slew processor. This is because the capacitor is probably being charged by an active circuit. On the S/H, the charge on the capacitor is "floating". _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Fri Oct 13, 2006 1:22 pm Post subject:
|
|
|
mosc wrote: | On the S/H, the charge on the capacitor is "floating". |
Yes, but I don't see how adding amplification would alter the circuit.
The idea is that the feedback circuit regulates the outpus in such a way that the negative input gets the same voltage as the positive input has. It does not matter here what the amplification is as long as the output does not clip. All input voltages and currents will be the same as in the non-amplyfying case.
So IMO, as long as the output does not clip, there can not possibly be any difference depending on the amplification of the buffer. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Fri Oct 13, 2006 1:31 pm Post subject:
|
|
|
Blue Hell wrote: | mosc wrote: | On the S/H, the charge on the capacitor is "floating". |
Yes, but I don't see how adding amplification would alter the circuit.
|
When you change the gain of the amplifier you change the input offset voltage of the op amp. It may not matter, but it could. The characteristics of the leakage of the charge on the capacitor will change as you change the gain. I would at least put a jfet or mosfet voltage follower in there. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Fri Oct 13, 2006 1:43 pm Post subject:
|
|
|
mosc wrote: | When you change the gain of the amplifier you change the input offset voltage of the op amp. |
As I reasoned all currents and voltages in/on the opamps inputs must essentially be the same (as long as the output doesn't clip), so I fail to see how the offset voltage could change. Could you explain this ? _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Fri Oct 13, 2006 2:32 pm Post subject:
|
|
|
That's not quite the point here (1), supposedly bad things were going to happen with the capacitor charge, I still don't see that.
(1) Indeed the offset voltage will be present on the output as an amplified signal - so stronger for less feedback (more gain). But that will happen in any opamp circuit and so adding a second independent gain stage will not improve things. No matter if its put before the holding capacitor or after the the buffer amp. And indeed in general it is not a good idea to first attenuate a signal and then amplify it again, but this was a design constraint here, certain circuit parts have to operate on a limited suppy voltage and so the attenuation has to be done. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
mosc
Site Admin
Joined: Jan 31, 2003 Posts: 18195 Location: Durham, NC
Audio files: 211
G2 patch files: 60
|
Posted: Sat Oct 14, 2006 9:17 am Post subject:
|
|
|
The effect can be subtile, but in keyboards even the tinyest error can cause unwanted and audible errors. An error of a millivolt could be a disaster.
I'm sorry I can't explain this adequately. When you change the operating parameters (such as gain, power supply, temperature) of a non-ideal op amp the ioc and voc change.
Do a search on sample and hold circuits and you'll see it is common practice to buffer the inputs and outputs with stable amplifiers.
http://www.du.edu/~etuttle/electron/elect53.htm
http://www.intersil.com/data/an/an517.pdf
Also, putting a pot in the feedback path of an op amp is to be avoided because pots are noisey and when put in the feedback path, the noise can be amplified, but that's another topic. _________________ --Howard
my music and other stuff |
|
Back to top
|
|
|
|