Bipolar PI work in progress

[Davor]

I think THAT 1570 would be a much better choice because it would enable you to supply it to a real balanced stage prior to the further processing. NE5534 ... is not a real difference amplifier because you do not have equal input impedances for differential input signal. A cross-coupled or Birt configurations would do that, yet neither of these, as well as your solution using a single NE5534 go down with noise.

There is a possibility to build a cost effective balanced CFIA solution using discrete BJTs in combination with cheap op amps. The same configuration is in THAT 1510, yet with not-too-good differential amplifier at the output.



In case you really wish to keeš NE5534, I'd suggest using NE5532 instead, but in a cross coupled configuration. It will provide balanced output which is good because you need reversed signal anyway.

[mickstv]

Hi Davor, thanks for the reply, I'll order one of the THAT1570's and have an experiment. I noticed that the THAT1510 made the output of the integrators more unstable for the same target response as the NE5534 very strange.

ATM I'm adding a diff amp to join the target and GB circuits, if it works ok I'll post up a modified circuit.

[Midas]

Hi Midas, for some reason when using the THAT1510 on high gains if I had larger targets near the coil the EF would become unbalanced, if I then ran the 1510 at a lower gain it cured the EF balance with bigger targets in the field but it lost response to smaller targets. I went back to the NE5534 because I knew it worked, Also if I decide to try the 1510 again I only have to move a couple of parts around to install it.

The timing previously posted was just a quick mock up. I have experimented with the GB samples at approx 40us width but still the same delay from the primary sample and this changed the detect hole. So eventually I'll have the processor setup to adjust the minimum decay target samples and decay + width of the GB channels.

Yes quicker pulsing does improve sensitivity.

Also I small typo on the circuit I actually run the circuit on about 8 volts.

Mick

Is it really a problem if the EF cancelation stops working when you have a big target near the coil ? I mean you have a signal anyway... unless of course the ground itself qualifies as a big target, then I see your problem.

Davor,
I have the 1570 here too.. I even have it on a carrier board. But looking to the future, its going to be a ***** to solder. Can't even use a wave tip on it, at least not if you want the underneath thermal\mechanical support pad connected, which I believe is highly reccommended. Still perhaps its worth the effort.

Midas

Hmm I wonder if dog breeding forums censor that word..

[mickstv]

Hi Midas, yes I would prefer the EF cancel to work perfectly and so far it does, using the the latest circuit shown.

Here is another mod, i've changed the output and have added another TL072 and moved the ground balance control. So now it's looking more like a GS5 circuit.

With the added IC the connections on the top CD4053 pins 10 and 11 have to be swapped.

[Davor]

IMHO your design is not suffering from EF (which doesn't make sense in bipolar build anyway), but from opamp input offset. You have over 5mV of offset with every FET input opamp that you simply can't compensate with common approaches. BJT input opamps on the other hand have excessive 1/f noise when compared to the FET ones in a same cost grade, but are perfectly compensated in a configuration shown in the schematic above. You may try NE5532 (BJT) instead of TL072 (FET), and see the difference in offset.

There is asymmetry in loading of the coils in this design because for a differential signal your classic diff amp provides about 1k1 input impedance at the inverting side, but 102.2k at noninverting. You can remedy that using either some of the balanced amps (THAT etc.) or some balanced pre - cross coupled amp, or Birt amp.

[mickstv]

Hi Davor, I'm working on the THAT ic again. I think your right it might be offset issues, but it still doesn't explain the fact that the circuit is working on the breadboard.

But anyway I'll try the NE5532's and see what it does to the output.


Cheers
Mick

[Janne]

There is asymmetry in loading of the coils in this design because for a differential signal your classic diff amp provides about 1k1 input impedance at the inverting side, but 102.2k at noninverting. You can remedy that using either some of the balanced amps (THAT etc.) or some balanced pre - cross coupled amp, or Birt amp.

Hi,

Could you elaborate what causes this difference in input impedances?

About the schematic, how can the feedback path of the primary amp work with just the 5.6p integrator cap?

[mickstv]

Hi,

Could you elaborate what causes this difference in input impedances?

About the schematic, how can the feedback path of the primary amp work with just the 5.6p integrator cap?

It's actually a mistake made by me, I missed a dot the 5.6pf should be connected to the 100k resistor on both sides and connected to pin 2 of the amp. This will be corrected in the next couple of days.



I should also mention to everyone this is a work in progress and as such will take a while to sort out all issues.



Cheers
Mick

[Davor]

A differential amp made with an op amp addresses several things simultaneously. It minimises offset in BJT input op amps, also it provides excellent common mode suppression with high and equal impedance for the common mode signal. It provides equal and opposite amplification for differential signal, but ONLY in case the inputs are hooked to a significantly lower impedance source, because the feedback reduces the input impedance of an inverting input.
There is a variant of a differential amp that provides symmetric load for differential signal, so called "true diff amp" but screws up the impedances of the common mode. So, using a single op amp you can't have both common and differential modes served correctly. No free lunch situation.

However, using 2 opamps in Birt configuration:


... or in cross coupled configuration (starred components are for floating output - just place ground at "centre tap" instead, 24R's at output can be omitted):


Of the two configurations I'd chose a cross coupled configuration because it has better high frequency behaviour than the Birt configuration. Both configurations provide equal impedance for both common and differential mode signals. Both can provide more gain than in the shown diagrams.

[Janne]

A differential amp ...

Thank you davor for your reply, the picture clears the idea for me now.

[kt315]

Janne, how are you there? what your Project stage is on now?

regards

[greylourie]

Janne, how are you there? what your Project stage is on now?

regards

Hi kt315, I think this was mickstv thread and work. Seems like he didn't carry on posting about it..... I wonder why. Maybe he got side-tracked by the work of another very clever guy, see here : http://www.geotech1.com/forums/showt...nduction/page2

[mickstv]

Hi kt315, I think this was mickstv thread and work. Seems like he didn't carry on posting about it..... I wonder why. Maybe he got side-tracked by the work of another very clever guy, see here : http://www.geotech1.com/forums/showt...nduction/page2

I did get sidetracked by some other projects. But this one is back on the bench at the moment. The design is completely different to what I originally posted and on the bench it seems to be running ok. After some more testing and when I get time I'll post some more info.

[6666]

Today I revisited a small bipolar TX project thats been on the list for a long time,
the progress looks promising, very simple idea for pulse generator, just a 556 .
Have not had time to tweek anything, but it ALMOST works. I plan to try it driving two fets @ 12v,with a 300uH centre tapped coil .

[Orbit]

may 4060 and this -450uh 4.3 khz -tx