Hi Daniel,
Thanks, so i can use OPA1642 for both locations. OPAs +ve --- cap --- ground OPA -ve --- cap --- ground X2 since changing 2 opamps
An externally hosted image should be here but it was not working when we last tested it.
oscillations in the opamp choice? or changing both, this is the picture imagine the cap pin is on the -ve and +ve of the opamp and earth is the black line from rca earth?
If you look at the circuit, you'll see a number of grounds. Now, I'm not 100% sure where the best ground is going to be on your pcb. In this case, might be best to join the legs of the grounded side of respective caps and then use a jumper wire to a low-impedance ground node (less than perfect).
Oscillations would be in the opamps for some reason. It happens when a circuit isn't deliberately designed for the application (and your bypass isn't close enough, but that's what we're trying to avoid).
Oscillations would be in the opamps for some reason. It happens when a circuit isn't deliberately designed for the application (and your bypass isn't close enough, but that's what we're trying to avoid).
There is a few grounding points on the pcb with copper traces, so i can solder cap pin to. Would a LME be less likely to oscillate than a OPA opamp. LME49860 seems to be well liked also but it isnt a jfet.
These caps on the opamps is to avoid oscillation hence findng a good ground for the cap pin, closest to opamp?
These caps on the opamps is to avoid oscillation hence findng a good ground for the cap pin, closest to opamp?
If anything, the 1642 should be less likely. It's the "slowest" opamp of that list overall.
Yes, that's a big reason (maintain low PSU impedance even in low-MHz range) for the caps. A simple star ground to a nearby ground is solid.
Yes, that's a big reason (maintain low PSU impedance even in low-MHz range) for the caps. A simple star ground to a nearby ground is solid.
Last edited:
Awesome Thanks Daniel, I shall order 2 OPA1642s as i have free shipping with Element14 no min buy 😛 only 7.38 each
The caps on the opas as in picture above with ground nearby Opamp
I had not noticed the OPA1642 prior to seeing it mentioned here. Looks like a winner!! A rail-to-rail JFET, very low noise and distortion, superb CMRR, dual SOIC, and even less $$$ than an OPA2134.
TI industriess sugguested it for my Technics SL-P990 which, has old Jfet Mitsubitshi opamps in it.
Even better would be a copper foil over the top of the opamp.
Connect the foil with two thin wires at each end into the PCB ground plane.
Finally bridge the small gap between the copper foil and each power pin with an smd ceramic just long enough to reach across.
The advantage of the copper foil is that it gives the return current passing through the decoupling capacitor the opportunity to take the lowest impedance path to get back to source.
Connect the foil with two thin wires at each end into the PCB ground plane.
Finally bridge the small gap between the copper foil and each power pin with an smd ceramic just long enough to reach across.
The advantage of the copper foil is that it gives the return current passing through the decoupling capacitor the opportunity to take the lowest impedance path to get back to source.
Not a good idea in my book. SMD ceramics are *extemely* prone to cracking off their terminations when subjected to seemingly "unproblematic" forces. "Dead-bug" wiring style of SMD caps and resistors is OK for testing on the bench but I would never recommend it for something that's actually in use.
Also, the extended ground plane over the IC, notably when it's SMD opamp, increases stray capacitance.
Those fast opamps that would actually benefit from a "shortest possible" decoupling path are also the most sensitive to excess stray capacitance.
Hi Andrew, Thanks. I would get copper ground planes from Farnell(Element14), as I think i have seen copper foil there.
Hi KSTR
Do you mean with the capacitor pins and to nearest ground?
The copper foil was shown in a short paper written by an IC specialist. I think he was an IC designer.
The foil in his example was to connect a single polarity power pin to ground on the diagonally opposite pin.
He explained that the foil needed a direct connection to the ground pin and should extend across the top to almost reach the power pin. That final gap was where he showed a small ceramic cap. This was quite an old paper and did not cover smd size IC, or components. It was a dip8, or dip14, in the photograph.
The foil in his example was to connect a single polarity power pin to ground on the diagonally opposite pin.
He explained that the foil needed a direct connection to the ground pin and should extend across the top to almost reach the power pin. That final gap was where he showed a small ceramic cap. This was quite an old paper and did not cover smd size IC, or components. It was a dip8, or dip14, in the photograph.
- Status
- Not open for further replies.