WELCOME!!

[Achillion]

Thanks for clarification. In this case drawing could be more consistent this way:

I've actually build this circuit but it draws quite some current and the LMs are getting hot also. I've constructed a circuit around a TLE2426 rail splitter and a LT1210 buffer which works much better. See the schematic attached below. The LT3042 can be ignored, you can include your own stabilized supply or go straight from the battery. In all cases the virtual ground will be kept at 1/2 of the supply voltage. The buffer gets warm but definitely draws less current that the dual regulator option.

[WM6]

No, they are connected as shown.

Thanks for clarification. In this case drawing could be more consistent this way:

[Old cart]

Thanks, this is not my original idea. I borrowed it from my other hobby-audio. It is used to run low noise audio circuits and can deliver up to 1.5 Amps.
No, they are connected as shown. In reality they are not absolutely required they are like using a suspenders on with your belt.

An even simpler design with a preregulator is shown here. (caps left off for clarity) This version uses fixed regulators(left) and adjustable pre regulators (right). The supply splitters have to be 1/2 the pre regulated voltage.
While this design is not quite as good as the previous one it may well be adequate for this application.

[WM6]

Very nice symmetric PS proposal Old Cart. Thanks.
Probably both 1N4002 are connected to ground too?

[Thomas]

Here is a front end that caught my eye recently.

[Old cart]

Oh it actually worked, just did not look that way! Anyway this supply is flexible. Easy to change supply voltage ranges etc. It basically splits whatever you give it so if you apply regulated 9 volt you get regulated +/-4.5 Volts. Since you would normally use a pre regulator of say 10 volts this would give a clean 10 volts to run transmitter driver stages and also be a clean source for the split supply. Not shown here is the extra bypassing that should be added, much as Sean has done. I think 100, 1 and .1uf ( film) should work well for C4 and C5. C1 Should be on the the order of 2200uF low ESR type. Coil should be fed with source as close as possible to this cap, star arrangement.

[Old cart]

Power supply

Ok I will do the power supply. Assuming we stay close to your original design I suggest we either we use the design you proposed or the following design.

I am having problems attaching an image, will try in next response.

This can be proceeded by a 11-15 volt battery and can produce any +/- voltages desired. Some samples are shown. The coil MOSFET will be driven either the battery directly OR the pre regulator that would feed this. The pre regulator is not critical and can be an ordinary 78 or 79 series TO220 style. By choosing the correct pre regulator I think this will work with either a N or P channel MOSFET. It should be able to source plenty of current, requires no adjustment, provides a very clean low impedance ground for the analog sections, and always give an perfectly symmetrical supply that tracks the input voltage.
I have NOT built this circuit but it is used in high end audio equipment where clean supplies are critical. It contains absolutely no switcher so could be built on the main board or as a separate module.

[Sean_Goddard]

You can find on places like thingyverse, generic opamp layouts/modules etc. Could be a useful way to learn about practical layout placement, and mixing manhattan and ugly style construction too.

Tell us more please. Can you investigate this for the project, in fact what about reference designs for op amps etc.

Invonic, What about the OP series amps, can you get those in DIP where you live? They are low noise and quite good.

[Sean_Goddard]

Right I'm back, here goes.

Two coil or mono or option? Let's go Mono to start and work from there.

Davor and Ivconic, can you take on the front end ampl please between you as a team as you BOTH have excellent ideas and I think that 1nv/Hz would be just about dandy ;-)). Keep the gain low so as to avoid ground problems and watch for saturation and recovery times.

Can you also look as active blanking or can we get away with not needing that? Suggestions please?

I'm going to task Old Cart with the PSU, nothing complicated, but rock solid stable and quiet.

I need someone to take on the integrators so I might have a go at that (EEK!)

Oscillator and pulse timing ANYONE please? Max of 2KHz and everything else variable too. One thing I want to try / ask about is a SUPER LONG ground integration time. Like can we sample the ground for 250uS and integrate it until JUST before the Tx pulse? Surely this would give us the best ground immunity or am I missing something?

Coil design, nothing hard to make just FAST (like 8uS) so we can get that thin yellow stuff.

Let me know if you're happy with these assignments.

[localbug]

I like the schematic, especially attention to supply filtering/bypass. And shift towards using 4066 or similar instead of fets for gating. And the I also like the thoughts on a "LEGO" type concept. Have thought about the idea too, as it would be easier to experiment with changes to a particular section or "block". And it would become very easy to create a model following the pattern of any older or well known design using generic "blocks" or modules. You can find on places like thingyverse, generic opamp layouts/modules etc. Could be a useful way to learn about practical layout placement, and mixing manhattan and ugly style construction too.

can't agree more for stack-up design, finally someone brought this to the surface.
will save plenty of time redesigning PI... power supply block, oscillators, front end, sampling, integrator,etc.
universal dimension need to be figured out..

[ivconic]

Not at all. You need to match a pair of FETs and that's about it. FETs are quite forgiving when it comes to matching.

Ah so. I though you are referring on special monolithic differential pair.
I saw those only on picture...

[Davor]

...Differential pairs are exotic thing here in local shops! Kinda "say what!!??" ...

Not at all. You need to match a pair of FETs and that's about it. FETs are quite forgiving when it comes to matching.

[Qiaozhi]

I don't want to be critical in any way , but I have seen many start-up projects on this forum , started and not finished, and when I saw suggestions like ... lets make it modular, add wireless charge .. I thought ooh here we go again ..Sean's projects will be lenghten again..and probably will end up like others ... which I really don't want to happen...

Hence my comment about the project engineer's hat.

The very valid comments you made (above) are nearly always true when something is designed by committee.

Sean - I would seriously suggest that you take complete charge of this project, and then it will be up to you if you include any of the suggestions or not. Otherwise, as soverey quite rightly said, the project will drag on forever with no ending in sight. Communal projects can seem a good idea at the start, but even with the best of intentions, it will inevitably come unravelled.

[ivconic]

With BF862 this circuit's noise may go below 1nV/sqrt(Hz)
Unfortunately it is also a bit misleading, because of a problematic settling time (those 1000uF capacitors), and the fact that this is a PI front end.
There are differential pair solutions that may prove better in settling time sense, and that may keep the good noise performance of a discrete front end.

Caps can be omitted or changed. Depends on power supply stage and how it is designed.
I put them only principally.
Differential pairs are exotic thing here in local shops! Kinda "say what!!??" reaction that you will get from a seller.
For guitar pickups is not problem to chose a fet. Here... i am not sure.
Yup, that's the misty part in schematic, i am scratching my head right now!

[Davor]

With BF862 this circuit's noise may go below 1nV/sqrt(Hz)
Unfortunately it is also a bit misleading, because of a problematic settling time (those 1000uF capacitors), and the fact that this is a PI front end.
There are differential pair solutions that may prove better in settling time sense, and that may keep the good noise performance of a discrete front end.