Ok. i'm bashing my head against the wall for long enough so i'm going to ask here.
Ive been experimenting with various amps both fully-solid state and hybrid on a breadboard, but when i start using opamps in combination with high-impedances (100k and higher) i notice that my experiments suddenly get very sensitive to noise from other stuff like dc/dc regulators and the likes, but i never seem to get fully rid of it. Ive tried multiple supply filters and eventually took a look through my college's equipment and they are pretty clean, but the noise remains. I keep hearing a high-pitched EEEEEEEE sound (yes i use grid-stoppers,high-pass filters and so on to prevent oscillation).
Eventually i discovered that this noise is most likely Ground related. Even small changes in how i wire everything together affects the noise for good or for bad, but even when i try to adhere as much as possible to Star-style grounding i get close...but never completely get rid of it.
So that brings me the question: Am i going insane?? or this is actually expected behaviour considering that i'm working on a Breadboard with (sometimes long) jumper wires and 100k-1M impedances (which equal tiny currents)?? And would the noise be less when i move a design from Breadboard to a PCB???
Ive been experimenting with various amps both fully-solid state and hybrid on a breadboard, but when i start using opamps in combination with high-impedances (100k and higher) i notice that my experiments suddenly get very sensitive to noise from other stuff like dc/dc regulators and the likes, but i never seem to get fully rid of it. Ive tried multiple supply filters and eventually took a look through my college's equipment and they are pretty clean, but the noise remains. I keep hearing a high-pitched EEEEEEEE sound (yes i use grid-stoppers,high-pass filters and so on to prevent oscillation).
Eventually i discovered that this noise is most likely Ground related. Even small changes in how i wire everything together affects the noise for good or for bad, but even when i try to adhere as much as possible to Star-style grounding i get close...but never completely get rid of it.
So that brings me the question: Am i going insane?? or this is actually expected behaviour considering that i'm working on a Breadboard with (sometimes long) jumper wires and 100k-1M impedances (which equal tiny currents)?? And would the noise be less when i move a design from Breadboard to a PCB???
Last edited:
I'll post this just for interest although it doesn't really solve your problem:
http://www.diyaudio.com/forums/solid-state/126370-oscillograms-testing-why-layout-matters.html
Yes, physical layout and the way signal returns and grounds are configured play a massive part. Also, high impedances and DC-DC convertors don't sound like a good combination tbh.
http://www.diyaudio.com/forums/solid-state/126370-oscillograms-testing-why-layout-matters.html
Yes, physical layout and the way signal returns and grounds are configured play a massive part. Also, high impedances and DC-DC convertors don't sound like a good combination tbh.
>100k Z with stripboard protos, op amp's source/feedback is asking for problems
the input parasitic C of the op amp's inputs can affect stability, transfer function, much less the larger stray C of the prototyping board contact strips to each other
dead bug on solid ground plane pcb is a good prototyping technique - Linear Technology showed SOTA performance of their chips with hand built protos
http://cds.linear.com/docs/en/application-note/an47fa.pdf
pth pad per hole with ground plane is another good base for prototyping
at high Z electric field coupling can become a problem, shielding can be built up with more pieces of clad pcb stock soldered into boxes
the input parasitic C of the op amp's inputs can affect stability, transfer function, much less the larger stray C of the prototyping board contact strips to each other
dead bug on solid ground plane pcb is a good prototyping technique - Linear Technology showed SOTA performance of their chips with hand built protos
http://cds.linear.com/docs/en/application-note/an47fa.pdf
pth pad per hole with ground plane is another good base for prototyping
at high Z electric field coupling can become a problem, shielding can be built up with more pieces of clad pcb stock soldered into boxes
There's a high probability that "a high-pitched EEEEEEEE sound" is coming from dc/dc regulators either through the rails of via EMF.
So, as already mentioned by jwilhelm and Mooly - PSU configuration, filtering, wiring, returns, grounding scheme - all these things are important.
Most likely, the breadboard itself is ok - the issue is somewhere around it.
So, as already mentioned by jwilhelm and Mooly - PSU configuration, filtering, wiring, returns, grounding scheme - all these things are important.
Most likely, the breadboard itself is ok - the issue is somewhere around it.
Your design will be much quieter in a properly designed PC board. The layout of the board, and the installation of the boards into a chassis are equally as important as the amplifier design itself.
Good to know i'm not going insane.
I'll post this just for interest although it doesn't really solve your problem:
http://www.diyaudio.com/forums/solid-state/126370-oscillograms-testing-why-layout-matters.html
Yes, physical layout and the way signal returns and grounds are configured play a massive part. Also, high impedances and DC-DC convertors don't sound like a good combination tbh.
Im mostly working with the stuff i have available to me. i dont have the resources for linear regulators with large heatsinks or transformers to get the voltages i need so i'm making due with smaller scale stuff including DC-DC regulators. gotta say it does bring up interesting obstacles.
dead bug on solid ground plane pcb is a good prototyping technique - Linear Technology showed SOTA performance of their chips with hand built protos
http://cds.linear.com/docs/en/application-note/an47fa.pdf
Awesome. whilst have seen a lot of Jim's other writings not seen that before.
mine works good no parasitic oscillations or anomalies IF you put a ground wire between each connection on the board so there are no two adjacent connections floating or otherwise ,all connections go to a circuit or go to ground no adjacent signals on protoboard.
do not put a metal Groundplane directly under the Proto board as this makes a capacitor on each connection because you have only a paper insulator label so put some type of space or cardboard or something between a metal groundplane and the Proto board so you don't build a capacitor to ground
do not put a metal Groundplane directly under the Proto board as this makes a capacitor on each connection because you have only a paper insulator label so put some type of space or cardboard or something between a metal groundplane and the Proto board so you don't build a capacitor to ground
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Solid State
- Breadboard, high-impedance and a LOT of noise.