yes, that is obvious... opamps have thd. SO, which LP filter topology will also reduce (filter) the opamp output as part of the filter topology?
THx-RNMarsh
THx-RNMarsh
only those with passive filters after the op amp output - at the price of higher and frequency variable Zout
Bruce Hofer that Scott mentioned had an idea on that.
Distortion reduction circuit for an inverting feedback amplifier
George
Distortion reduction circuit for an inverting feedback amplifier
George
Is there a reason why nobody mentions the SV filter with the output after an integrator? Or am I giving away the game? 😱
Jan
Jan
http://www.brl.uiuc.edu/Downloads/Embree-MS-1982/paulEmbree-MS-Thesis.pdf
This is one paper that actually mentions distortion in SV filters.
This is one paper that actually mentions distortion in SV filters.
Here is a tool for multiple feedback filter design Multiple Feedback Low-pass Filter Design Tool The issue with that filter is noise. Also true of any inverting amplifier.
The classic tradeoff, the same is true of input Z matching you get at least a 3db hit on source impedance noise.
Also to try --- Anyone know of a computer design using all passive parts of a high slope rate atten LP filter? at least 24db/oct or greater.
THx
RNMarsh
THx
RNMarsh
Problem for me is the LP output will be the thd of the opamp. I am looking for topology which will have lower thd at its output than the opamp used. That is, the ouput for the Lp filter would also be filtered inherently by the topology.
THx-
RNMarsh
The unity gain distortion of the better opamps is extremely low. A passive filter on the output may lower whatever harmonics are still present. How would you tell if there were harmonics? Typically the closed loop distortions are well below the best analyzers today.
How? Build the filter with LM358s {opamps with ferocious amounts of distortion} and see whether the filter output has lower harmonics than the LM358's native harmonics. If so, you have good reason to believe the filter topology gives output harmonics less than opamp harmonics. Swap in SOTA opamp and be happy, even if you can't measure the improvement any more.
Hey, John, The book came in the mail today, such as this:
ACTIVE-FILTER
COOKBOOK
DON LANCASTER
It is a Howard W. Sams & Co., Inc. Book.
And now I see there is a name inside:
Mark E Johnson
F1J1
It's the copyright 1975 first edition, seventh printing from 1981.
Mark, is that one of your old books or just coincidence?
I've perused it and it looks to be a keeper.
I pulled another book I started studying before
Lancaster's book showed up and it discusses what
Ed and some other folks were discussing here.
Namely, there are some losses, paracitics and related
issues that aren't readily solveable by simply putting
High-Pass Filter and Low-Pass Filter connected in series.
I suspect Lancaster discusses it but I haven't found it yet.
I can post a copy of the paper here if anyone is interested
or actually send it. I'd post a copy but I think the blogs
here have a size limit. Sadly, or not, the calculations where
done the old fashioned way, then I took a pic, made them
into jpeg and inserted them into the paper. So the data is
there, just not all typed p in a table, etc.
Okay, gotta run, cheers,
COOKBOOK
DON LANCASTER
It is a Howard W. Sams & Co., Inc. Book.
And now I see there is a name inside:
Mark E Johnson
F1J1
It's the copyright 1975 first edition, seventh printing from 1981.
Mark, is that one of your old books or just coincidence?
I've perused it and it looks to be a keeper.
I pulled another book I started studying before
Lancaster's book showed up and it discusses what
Ed and some other folks were discussing here.
Namely, there are some losses, paracitics and related
issues that aren't readily solveable by simply putting
High-Pass Filter and Low-Pass Filter connected in series.
I suspect Lancaster discusses it but I haven't found it yet.
I can post a copy of the paper here if anyone is interested
or actually send it. I'd post a copy but I think the blogs
here have a size limit. Sadly, or not, the calculations where
done the old fashioned way, then I took a pic, made them
into jpeg and inserted them into the paper. So the data is
there, just not all typed p in a table, etc.
Okay, gotta run, cheers,
Well I just got this years spam award.
"Edward Simon & Company. has been selected for the 2017 Best of Pittsburgh Awards for Water."
I did find out why it is for water. Google maps lists my location as a body of water! Before they built the nearby interstate in the 50's they had to divert a stream. What was once water is now my front parking lot. My address used to be my back door. When I bought the building I installed a door in the front to the parking lot. This changed my mailing address!
I also got one for Edward Simon & Co. being the best electronics business in my actual location of Sharpsburg. A huge honor as there are no other ones in town. 🙂
My favorite mistaken advertisement was years ago in a yellow pages spin-off where the bullet of "Free Sound Surveys" became "Free Sound Systems."
RNM,
I suspect but have not run the numbers that an RC filter after an op-amp based one would lower the distortion by the expected 6 dB per octave but due to the output load requirements would raise the noise. In the past folks did sell multi-pole LC filters. Besides the issue with inductors I am sure you are aware of, they also required specific input and output impedances to work correctly. My approach would be to parallel op-amp based filters and then combine them through resistors into a single capacitor. I just can't think of any better approach.
"Edward Simon & Company. has been selected for the 2017 Best of Pittsburgh Awards for Water."
I did find out why it is for water. Google maps lists my location as a body of water! Before they built the nearby interstate in the 50's they had to divert a stream. What was once water is now my front parking lot. My address used to be my back door. When I bought the building I installed a door in the front to the parking lot. This changed my mailing address!
I also got one for Edward Simon & Co. being the best electronics business in my actual location of Sharpsburg. A huge honor as there are no other ones in town. 🙂
My favorite mistaken advertisement was years ago in a yellow pages spin-off where the bullet of "Free Sound Surveys" became "Free Sound Systems."
RNM,
I suspect but have not run the numbers that an RC filter after an op-amp based one would lower the distortion by the expected 6 dB per octave but due to the output load requirements would raise the noise. In the past folks did sell multi-pole LC filters. Besides the issue with inductors I am sure you are aware of, they also required specific input and output impedances to work correctly. My approach would be to parallel op-amp based filters and then combine them through resistors into a single capacitor. I just can't think of any better approach.
@ Ed - That's pretty funny Ed. I wish I would have known,
I would liked to have had a free sound system though.
Quite the accomplishment and shows a bit of humility also,
with the Best ELE shop. 🙂
Also, I'm glad you work on boats Ed, did you know I was about
to schedule the Water Ski World Championships in your parking lot.
Thank you for the clarification...I'm sure ABCs Wide World of Sports
would have had a field day on the production aspects of it.
Then there was the guy who was changing his flight from SFO - ORD
he went to the counter and the lady behind it was extremely well endoud,
he blurted out, I need to change these, "I need two pickets to titsburgh."
@ Mark, one never knows until they ask. I might have thought
you were liquidating your design library and was going to ask
for a discount if I took them all.
Quasimodo - Good pics with better callouts that mine, as I tried to put
a background box around my text but program wouldn't'cooperate.
Is that Mark Q. Johnson?
Cheers,
I would liked to have had a free sound system though.
Quite the accomplishment and shows a bit of humility also,
with the Best ELE shop. 🙂
Also, I'm glad you work on boats Ed, did you know I was about
to schedule the Water Ski World Championships in your parking lot.
Thank you for the clarification...I'm sure ABCs Wide World of Sports
would have had a field day on the production aspects of it.
Then there was the guy who was changing his flight from SFO - ORD
he went to the counter and the lady behind it was extremely well endoud,
he blurted out, I need to change these, "I need two pickets to titsburgh."
@ Mark, one never knows until they ask. I might have thought
you were liquidating your design library and was going to ask
for a discount if I took them all.
Quasimodo - Good pics with better callouts that mine, as I tried to put
a background box around my text but program wouldn't'cooperate.
Is that Mark Q. Johnson?
Cheers,
Last edited:
As I said, a few pages back. Nobody picked it up. You can pick off the output of a SV after the integrator which of course decreases the harmonics.
Jan
x2
This is very much an application-specific analysis -- need to have a clear idea of the impedances floating around and what you're hoping to accomplish, and would need to be characterized in-place (or thorough simulation with GOOD models) to see if the loading doesn't completely throw away any ostensible distortion benefit. Paralleling devices comes with its own set of compromises, although quite manageable from line-level.
- Status
- Not open for further replies.