field test unit no 001 "model T"

[Midas]

Yeah sorry mick, that's not what I meant but it's my fault, I was explaining my confusion poorly. I've got it sorted now, for some reason I had it in my head that the ref connection on the instrumentation amplifier arrangement was the actual gnd pin of the op-amp, pretty much just because it went to ground and the schematic omitted the actual gnd. No consideration was given to the fact that it clearly goes into the non-inverting input... pretty poorly thought out really... I think I'll drink less vodka tonight.

[mickstv]

The inputs are high ie:- 12 volts, as long as the opamp doesn't have any reference to ground (no resistors or capacitors going to ground) it will see the voltage rails as being +/- 5volts. Thats how I see it anyway.



Mick

[Midas]

Using your multimeter put the negative black lead on 12 volts then you'll see + and - 5 volts out of the two regulators.



Mick

I understand the concept of the negative supply, but where does it actually hook into the instrumentation amplifier op-amp arrangement? The THAT 1510 has a reference and a negative supply pin. Op-amps like the LM324 don't..

[mickstv]

I'm still confused. Ref is 12v logically enough but where does the negative supply (+7v in the schematic, pin 4 on the THAT) connect up to ?

Using your multimeter put the negative black lead on 12 volts then you'll see + and - 5 volts out of the two regulators.



Mick

[Midas]

Opamp instrumentation amp connection

Connect the LM324 diff amp equivalent in the same manner. The reason the ref goes to "positive" is because this is the analogue ground ( centre tap of coil ).


moodz.

I'm still confused. Ref is 12v logically enough but where does the negative supply (+7v in the schematic, pin 4 on the THAT) connect up to ?

[6666]

Connect the LM324 diff amp equivalent in the same manner. The reason the ref goes to "positive" is because this is the analogue ground ( centre tap of coil ).


moodz.

Thanks I see the light

( centre tap of coil ).

[moodz]

Looking at the general block diagram of the THAT chip
do you think the lm324 amp would be ok for experimental purposes ?
with a pot for the gain

The only thing I am not sure of is the ref connection which is pin 5 on the THAT

the ref connection seems to equate to the positive input of the last lm324


Moodz has it going to positive, but the lm324 cct has it going to gnd
which should it be for our cct ?

Connect the LM324 diff amp equivalent in the same manner. The reason the ref goes to "positive" is because this is the analogue ground ( centre tap of coil ).


moodz.

[6666]

Looking at the general block diagram of the THAT chip
do you think the lm324 amp would be ok for experimental purposes ?
with a pot for the gain

The only thing I am not sure of is the ref connection which is pin 5 on the THAT

the ref connection seems to equate to the positive input of the last lm324


Moodz has it going to positive, but the lm324 cct has it going to gnd
which should it be for our cct ?

[moodz]

The chip I use is CD4066BCN and the supply rail is set to 17 volts ( 12 + 5 ) and logic ground. In this circuit the analogue ground is the voltage at the "centre tap" of the coil.
The amplifier +/- 5 volt supply is referenced to the analogue ground.

CD4066BCN spec is ....
Wide supply voltage range 3V to 15V
input range is +/- 7.5 volts peak

the spec sheet should really say input is +/- vsupply/2

you can run the chip on more than 15 volts ...this spec is for CMOS logic compliance however we are not using chip for logic but analogue switching

moodz.

[Vortxrex]

FETs now balanced NP on each input.
NP Mosfet pack used ( 300 volts .... but good for 350 )
4066 works better than other more expensive solutions but use mosfet switch to drive ( like a tc4421 ) Make sure you use the high voltage 4066 ( eg 15 volt rating ).
The THAT amplifier is not the best here ... use opamp DC diff amp.

moodz.

[ATTACH]17470[/ATTACH]

[Midas]

Hi Moodz, on IC4 4066n are you supplying voltage from the +/- 5volt regs or is it getting supply from the VPP 12 volts ?



Mick

Hi Mick,

From the 4066 datasheet:
Input voltage range, Vis (all inputs) . . . . . . . . . . −0.5 V to VDD + 0.5 V

The positive would have to be +5v since Vin shouldn't exeed VDD by more than .5v with the 4066 (Moodz has already demonstrated getting over 1v of un-amplified signal with some targets). The negative connection would have to be actual ground by the same logic, since if you use the -5v then the Vin would go too far negative during transmit where one side is grounded.

Sorry I know you weren't asking me, but Moodz seems a little busy lately.

Regards
Midas

Edit:
Actually it should be fine to use the -5v, the fets should block the signal from going too low during transmit.

Second edit: ( Mental note, engage brain before keyboard ) It actually looks like the fets may keep the signal below .5v which would possibly would allow you to use the 12v..
You should probably ask Moodz. LoL..

[mickstv]

Hi Moodz, on IC4 4066n are you supplying voltage from the +/- 5volt regs or is it getting supply from the VPP 12 volts ?



Mick

[Midas]

Most of the drift comes is contributed by C4 and D5 .... the charge on C4 is not deterministic due to ground and large target effects ....that is why I changed to a separate bias generator and floating ground in the latest iteration of the design.

moodz.

Yeah right that's not really drift though, its motoring in the wrong direction
But anyway, on the bench, without moving the coil at all, are you happy with the stability? I guess a bit of slow drift isn't necessarily a problem anyway. Although it may force the design into the motion required detector category (or is it there already?).

[moodz]

Yeah I doubt my circuit will be any better in terms of drift in any case.
But are you sure its not the mosfet damping causing the drift ? What are your other contenders? Because I would expect the gate voltage to resistance relationship of a mosfet to be somewhat temperature dependent (shame there's no direct information on this in the datasheet) so without any feedback or temperature compensation I would think it would be impossible to eliminate drift completely. Easy test for you anyway, try freeze spraying your damping mosfet. That will prove that it is a source of drift, though I guess not necessarily that's its the dominant source under normal conditions.

Most of the drift comes is contributed by C4 and D5 .... the charge on C4 is not deterministic due to ground and large target effects ....that is why I changed to a separate bias generator and floating ground in the latest iteration of the design.

moodz.

[Midas]

Drift musings

It could work .. I will probably play more with the damping once the rest of the detector is put together. At the moment the existing damping circuit can balance the detector even at extremely high gains ( 1000 - 2000 x ) on the diff amp.... however it drifts but that is for other reasons ... not the damping.

moodz.

Yeah I doubt my circuit will be any better in terms of drift in any case.
But are you sure its not the mosfet damping causing the drift ? What are your other contenders? Because I would expect the gate voltage to resistance relationship of a mosfet to be somewhat temperature dependent (shame there's no direct information on this in the datasheet) so without any feedback or temperature compensation I would think it would be impossible to eliminate drift completely. Easy test for you anyway, try freeze spraying your damping mosfet. That will prove that it is a source of drift, though I guess not necessarily that's its the dominant source under normal conditions.