For some time I've been thinking about doing an attenuator based on an I/V converter.
When doing some web searches on the idea, I came across this paper that some of you may find interesting.
http://www.esscirc.org/esscirc2000/proceedings/pdf/10.pdf
regards, jonathan carr
When doing some web searches on the idea, I came across this paper that some of you may find interesting.
http://www.esscirc.org/esscirc2000/proceedings/pdf/10.pdf
regards, jonathan carr
I don't know about that 1% distortion spec.
You could always use a D/A multiplier circuit like the good
folks at Madrigal, probably getting better than that.
Then again, maybe not...
😉
You could always use a D/A multiplier circuit like the good
folks at Madrigal, probably getting better than that.
Then again, maybe not...
😉
Nelson: It is possible that the "audiometric" circuit simply isn't suited for low distortion, but then I haven't had a chance to even analyze the paper fully, let alone try to come up with my own twists on it.
But I have done some work with discrete Gilbert multipliers as attenuators, and I felt that the two issues there were DC offset (which I figured out how to deal with), and lack of dynamic range (which I haven't figured out how to address). At least the present "audiometric" design doesn't appear to have ANY problems with dynamic range, hence my interest. At least I hope that it will be able to jar loose an inspiration or two.
I believe that Japan Victor Coporation (JVC) has a different complementary I/V-based attenuator circuit which they have used in some of their integrated amplifiers and preamps. This is supposed to be quite good, but I haven't managed to unearth the patent numbers yet.
regards, jonathan carr
But I have done some work with discrete Gilbert multipliers as attenuators, and I felt that the two issues there were DC offset (which I figured out how to deal with), and lack of dynamic range (which I haven't figured out how to address). At least the present "audiometric" design doesn't appear to have ANY problems with dynamic range, hence my interest. At least I hope that it will be able to jar loose an inspiration or two.
I believe that Japan Victor Coporation (JVC) has a different complementary I/V-based attenuator circuit which they have used in some of their integrated amplifiers and preamps. This is supposed to be quite good, but I haven't managed to unearth the patent numbers yet.
regards, jonathan carr
Of course the I/V part is after the attenuator, and what we
see are circuits operating into a virtual ground, which is
mostly how they get the distortion and dynamic range
that they do manage.
You can probably get better numbers with the higher quality
analog gates that are on the market now, placing the
gate just at the virtual ground so that it doesn't suffer from
voltage dependant resistance. This gets the distortion way
down and allows you to pump a fair amount of voltage at
the circuit. The "discrete" analog gates have the advantage
of more silicon area, and this makes them quieter and more
linear.
We have a (pat pend) circuit which uses bipolar elements to
shunt signal, and we get voltage, distortion, and noise about
1/10 the industry figures. You'd think the PO would get off
their butt and grant it....
see are circuits operating into a virtual ground, which is
mostly how they get the distortion and dynamic range
that they do manage.
You can probably get better numbers with the higher quality
analog gates that are on the market now, placing the
gate just at the virtual ground so that it doesn't suffer from
voltage dependant resistance. This gets the distortion way
down and allows you to pump a fair amount of voltage at
the circuit. The "discrete" analog gates have the advantage
of more silicon area, and this makes them quieter and more
linear.
We have a (pat pend) circuit which uses bipolar elements to
shunt signal, and we get voltage, distortion, and noise about
1/10 the industry figures. You'd think the PO would get off
their butt and grant it....
Ah yes, the PO debacle. Sympathy to you Nelson. One of recent took 4 years of arguing, while a competing firm was granted a patent on the size of a bearing used for a certain application. This would be like granting exclusivity to IRF-610's in one of your circuits. The system is just not right.
Charles BL
Charles BL
This is the closest JVC patent that I have been able to find. My memory is that the "Gm volume" that JVC was actually using is rather different from this topology. (I think that it was more like a complementary folded-cascode variant), but I haven't been able to locate any solid documentation.
http://patft.uspto.gov/netacgi/nph-...,169,247.WKU.&OS=PN/4,169,247&RS=PN/4,169,247
regards, jonathan carr
http://patft.uspto.gov/netacgi/nph-...,169,247.WKU.&OS=PN/4,169,247&RS=PN/4,169,247
regards, jonathan carr
atten etc
Jonathan,
Are you aware of the ADI series of "X-amp" VCA's like the dual AD604? A pair of cascoded amps gives close to 100dB dyn range, with very little distortion. Basical;ly its a diff input R/1.5R att followed by a .6nV/RtHz amp. The x-amp does lin interpolation between steps to get a lin-in-dB att curve. Worth a look if you're in that sort of thing.
Jan Didden
Jonathan,
Are you aware of the ADI series of "X-amp" VCA's like the dual AD604? A pair of cascoded amps gives close to 100dB dyn range, with very little distortion. Basical;ly its a diff input R/1.5R att followed by a .6nV/RtHz amp. The x-amp does lin interpolation between steps to get a lin-in-dB att curve. Worth a look if you're in that sort of thing.
Jan Didden
Actually, it's Wayne's patent that's hung up. I have stoppedCharles BL said:
collecting patents.
😉
Jan: Thank you for the notice.
I am primarily interested in designing a discrete circuit rather than use an IC, and besides, there are specific usage requirements and performance parameters about the AD604 that make me question how suitable it will be for my requirements.
Nonetheless, the data sheet appears to be an interesting read, and I hope that I will learn something useful from it.
Once again, thank you for letting me know.
regards, jonathan carr
I am primarily interested in designing a discrete circuit rather than use an IC, and besides, there are specific usage requirements and performance parameters about the AD604 that make me question how suitable it will be for my requirements.
Nonetheless, the data sheet appears to be an interesting read, and I hope that I will learn something useful from it.
Once again, thank you for letting me know.
regards, jonathan carr
- Status
- Not open for further replies.