4 of 4 people found this helpful
I think it's fine, although generally there is separation between the amplifier circuit, and the PSU, i.e. two separate ground planes, connected at one point. And as unbroken as possible (harder on a 2-layer board). A ferrite bead between the PSU and the amplifier circuit is useful too, not just to provide some high frequency filtering, but also so that if desoldered, the two circuits are isolated for testing. C6 and and C8 may need to be higher value than 10uF for good decoupling, so it could be worth leaving a slightly bigger space for them on the PCB.
Regarding track widths, generally I'll default to 8mil or 6mil for signals, and thicker (e.g. 16 or 24 or higher) otherwise. There's some standards for that, also some online PCB calculators, e.g. this one: https://www.4pcb.com/trace-width-calculator.html
Many thanks Shabaz,
The PCB track calculator is very useful, a link worth saving.
I will have a look to see if I can produce two ground planes. I may be able to join them at the corner of the transformer somewhere near R4, although the track through to R7 may become an issue. I think there is plenty of room to move the components around to split up the ground planes.
I will then look to see what ferrite beads would be available to me, I presume I would need on in each rail of the supply? The sizes for capacitors C6 and C8 came from the data sheet for the amplifier. I will also only look to go up to 30kHz with this as well as that is the maximum bandwidth of the transient analyser.
3 of 3 people found this helpful
I have carried out some more work on the PCB. As long as I have understood what you have meant.....
Ferrite beads have been added into each supply rail and I have juggled the components about to allow the ground planes to be split with just one connection between them. Not sure about what size that connection should have been, so I went with the electrical approach.
I am building up a little current amplifier to allow me to test the frequency response of some Rogowski coils that I want to adapt for use with a transient analyser.
The amplifier will be driven from a waveform generator and have a selectable 1A / 2A current limit on the output. This will drive a current clamp table I will build to give the higher currents for the Rogowski coils.
The PCB houses a simple dual rail power supply to provide power to the OP-amp, that I have created a large exclusion zone around the mains connections and tracks.
Below is the schematic, and the front and back copper layouts.
Two questions spring to mind.
1) The copper pour has been connected to the centre (0V) rail of the rectifier D1. Is this the best connection for it?
2) PCB tracks for mains, DC supply rails and amplifier output and 1.0mm wide tracks. Is this adequate enough?