Low voltage negative resistance oscillator design, lessons l

[Steve Wilson]

piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 10:03 am, Steve Wilson wrote:
Your help has been invaluable. The patent especially. Now we have
enough information to spend hours studying this amazing design. It
still has merit for the unmatched common mode range.

The AD310 common mode range is only good at DC/low-AF since they
penny-pinched and used capacitive coupling from the bridge to the AC
amplifier. A fully transformer isolated design like the Philbrick P2
would be good to much higher volatges and/or frequencies.

piglet

The 310 bridge is transformer coupled. Essentially unlimited common mode
range.

Why do you need wide bandwidth in a circuit intended for pA
applications?

Hi Steve,

The AD310 has transformer coupling from carrier oscillator to bridge,
but from bridge to AC amplifier is capacitor coupled. A glass 22pF ultra
low leakage cap in one leg and a garden variety ceramic 1nF in the other
(cold) leg. I suggest that a balanced transformer design like P2 or P2A
or the AD patent would have better AC CMRR.

Wide signal bandwidth is not the issue but AC common mode interference
could be a problem. Imagine the AC CMRR needed for a worst case dc-like
pA signal riding on 120V 60Hz !

piglet

You are not going to find pA signals riding on 120V 60Hz. The 310 was
intended for electrometer applications. These are static, DC applications.

In a related topic, whit3rd stated on Fri, 30 Aug :

"The AD310 usually sat in a teflon-bushed socket and might have to be
hand-wired,"

Page 1 of the AD 310 description talks about the small size, high input
impedance and low bias current needed for use in electrometer applications:

https://tinyurl.com/yyxlqtg8

Yet the pcb photo and schematic posted earlier shows a circuit that would
not fit in a teflon socket.

Which one is the real AD 310?

 

[George Herold]

On Tuesday, September 3, 2019 at 3:48:47 PM UTC-4, Blyat Cheker wrote:

On Tuesday, September 3, 2019 at 2:36:52 PM UTC+2, piglet wrote:
On 03/09/2019 9:51 am, piglet wrote:
I am puzzled how the oscillator works (those 2 pnps) - that is going to
be my coffee-time challenge.


I think those three things that like resistors "brown, gray, brown,
white" and "brown, black, brown, white" are not resistors but capacitors
of bzw 180pF and 100pF.

The curious looking oscillator then is nothing more than a usual astable
multivibrator.

piglet

I've appended the imgur post with a scope capture of the oscillator. It runs at about 200kHz.

Thanks that's interesting. I made this 'audio' double balanced mixer with
four diodes in a ring. I found that diode connected transistors (B-C shorted)
worked 'better' than regular signal diodes.

George H.

I do still wonder how they found the 2N5962. The datasheet says they are general-purpose amplifier transistors. The ones in the varactor amp are marked with red paint, too (a bit hard to see in the pictures), so AD must've handselected them.
Maybe they were just cheap? I guess the main important thing is the matching.

Nice info regarding the color codes and the capacitors "disguised" as resistors, I didn't know about that.

 

On Wed, 4 Sep 2019 06:37:37 -0700 (PDT), George Herold
<gherold@teachspin.com> wrote:

On Tuesday, September 3, 2019 at 3:48:47 PM UTC-4, Blyat Cheker wrote:
On Tuesday, September 3, 2019 at 2:36:52 PM UTC+2, piglet wrote:
On 03/09/2019 9:51 am, piglet wrote:
I am puzzled how the oscillator works (those 2 pnps) - that is going to
be my coffee-time challenge.


I think those three things that like resistors "brown, gray, brown,
white" and "brown, black, brown, white" are not resistors but capacitors
of bzw 180pF and 100pF.

The curious looking oscillator then is nothing more than a usual astable
multivibrator.

piglet

I've appended the imgur post with a scope capture of the oscillator. It runs at about 200kHz.

Thanks that's interesting. I made this 'audio' double balanced mixer with
four diodes in a ring. I found that diode connected transistors (B-C shorted)
worked 'better' than regular signal diodes.

The other connection, b-e shorted, makes superb diodes with higher
reverse voltage. I guess a bit of reverse beta makes the "forward"
conduction a little better.

A BFT25 is a better diode than practically any official diode.

 

[Steve Wilson]

j.ponte@student.utwente.nl wrote:

On Monday, September 2, 2019 at 8:33:48 PM UTC+2, Steve Wilson wrote:
whit3rd <whit3rd@gmail.com> wrote:

On Friday, August 30, 2019 at 4:39:30 PM UTC-7, Clifford Heath wrote:
On 9/8/19 2:52 am, bitrex wrote: ather than varactors.

It's a curious thought though, using a varactor as an amplifier.
Does anyone have an example circuit?

The Philbrick (P2?) was earlier, but this one is well documented

https://archive.org/details/AnalogDevicesDataAcquisitionProductsCatal
og1 979/page/n97

The AD310 usually sat in a teflon-bushed socket and might have to be
hand-wired, but it gave the vacuum-tube vibrating reed electrometers
some real solid state competition on leakage current. After all, if
you bias a varactor at zero volts, what IS the expected DC current?

I am trying to understand how this worked. There is a diagram of the
input circuit on page 3 of the file. Do you have any idea of the
frequency and amplitude of the signal fed to the bridge?

I did a teardown of one and reverse-engineered it:
https://imgur.com/jRxgMy4

What did you tear down and where did you get it?

According to whit3rd, the AD 310 sat in a teflon bushed socket. Your
circuit would definitely not fit in a teflon socket.

(Still trying to make sense of this thing

Thanks

 

[Tauno Voipio]

On 4.9.19 16:37, George Herold wrote:

Thanks that's interesting. I made this 'audio' double balanced mixer with
four diodes in a ring. I found that diode connected transistors (B-C shorted)
worked 'better' than regular signal diodes.

George H.

Half a century ago they were called transdiodes. The idea is that
the current gain of the transistor helps suppressing the effect
of the base resistance, and the transdiodes followed better the
theoretical diode equation.

--

-TV

 

[Piglet]

On 04/09/2019 16:57, Steve Wilson wrote:

j.ponte@student.utwente.nl wrote:

On Monday, September 2, 2019 at 8:33:48 PM UTC+2, Steve Wilson wrote:
whit3rd <whit3rd@gmail.com> wrote:

On Friday, August 30, 2019 at 4:39:30 PM UTC-7, Clifford Heath wrote:
On 9/8/19 2:52 am, bitrex wrote: ather than varactors.

It's a curious thought though, using a varactor as an amplifier.
Does anyone have an example circuit?

The Philbrick (P2?) was earlier, but this one is well documented

https://archive.org/details/AnalogDevicesDataAcquisitionProductsCatal
og1 979/page/n97

The AD310 usually sat in a teflon-bushed socket and might have to be
hand-wired, but it gave the vacuum-tube vibrating reed electrometers
some real solid state competition on leakage current. After all, if
you bias a varactor at zero volts, what IS the expected DC current?

I am trying to understand how this worked. There is a diagram of the
input circuit on page 3 of the file. Do you have any idea of the
frequency and amplitude of the signal fed to the bridge?

I did a teardown of one and reverse-engineered it:
https://imgur.com/jRxgMy4

What did you tear down and where did you get it?

According to whit3rd, the AD 310 sat in a teflon bushed socket. Your
circuit would definitely not fit in a teflon socket.

(Still trying to make sense of this thing

Thanks

The teardown AD310 in the wonderful photos had been soldered into the
wider world judging by the tinned tips of the tall pins. However page 98
of the AD catalog at

<https://ia801606.us.archive.org/35/items/AnalogDevicesDataAcquisitionProductsCatalog1979/AnalogDevicesDataAcquisitionProducts1979_text.pdf>

does describe a mating socket AC1017 that users could elect to use
instead of direct soldering. I guess the two input sockets of that
AC1017 would have teflon bushings. Teflon wouldnt have been necessary on
all the other i/o pins so it is possible those were insulated less
extravagantly. In the teardown photos one can clearly see the "hot" hi-Z
input circuit internally is on teflon standoffs so it makes sense to
keep up the high grade insulation.

piglet

 

[Phil Hobbs]

On 9/4/19 1:26 PM, Phil Hobbs wrote:

On 9/4/19 12:39 PM, jlarkin@highlandsniptechnology.com wrote:
On Wed, 4 Sep 2019 06:37:37 -0700 (PDT), George Herold
gherold@teachspin.com> wrote:

On Tuesday, September 3, 2019 at 3:48:47 PM UTC-4, Blyat Cheker wrote:
On Tuesday, September 3, 2019 at 2:36:52 PM UTC+2, piglet wrote:
On 03/09/2019 9:51 am, piglet wrote:
I am puzzled how the oscillator works (those 2 pnps) - that is
going to
be my coffee-time challenge.


I think those three things that like resistors "brown, gray, brown,
white" and "brown, black, brown, white" are not resistors but
capacitors
of bzw 180pF and 100pF.

The curious looking oscillator then is nothing more than a usual
astable
multivibrator.

piglet

I've appended the imgur post with a scope capture of the oscillator.
It runs at about 200kHz.

Thanks that's interesting.  I made this 'audio' double balanced mixer
with
four diodes in a ring.  I found that diode connected transistors (B-C
shorted)
worked 'better' than regular signal diodes.


The other connection, b-e shorted, makes superb diodes with higher
reverse voltage. I guess a bit of reverse beta makes the "forward"
conduction a little better.

Never tried that--good idea.  The Philips model has

BR=16.9 for the

BFT25A vs. BF=85, fwiw.


A BFT25 is a better diode than practically any official diode.

Yup, an old fave for that.  One of our products uses a small InGaAs
photodiode with a TIA whose feedback is 200M parallelled with two BFT25A
C-B junctions in series, with a feedback T-network that preserves the
full 5-V output range.  It's for a more qualitative sort of application,
so we don't care much about the diode tempco.  That combination gives
high sensitivity and constant gain in low light, and in brighter light
it preserves the the highest sensitivity consistent with not railing the
amp.

FTFM.

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Phil Hobbs]

On 9/4/19 12:39 PM, jlarkin@highlandsniptechnology.com wrote:

On Wed, 4 Sep 2019 06:37:37 -0700 (PDT), George Herold
gherold@teachspin.com> wrote:

On Tuesday, September 3, 2019 at 3:48:47 PM UTC-4, Blyat Cheker wrote:
On Tuesday, September 3, 2019 at 2:36:52 PM UTC+2, piglet wrote:
On 03/09/2019 9:51 am, piglet wrote:
I am puzzled how the oscillator works (those 2 pnps) - that is going to
be my coffee-time challenge.


I think those three things that like resistors "brown, gray, brown,
white" and "brown, black, brown, white" are not resistors but capacitors
of bzw 180pF and 100pF.

The curious looking oscillator then is nothing more than a usual astable
multivibrator.

piglet

I've appended the imgur post with a scope capture of the oscillator. It runs at about 200kHz.

Thanks that's interesting. I made this 'audio' double balanced mixer with
four diodes in a ring. I found that diode connected transistors (B-C shorted)
worked 'better' than regular signal diodes.


The other connection, b-e shorted, makes superb diodes with higher
reverse voltage. I guess a bit of reverse beta makes the "forward"
conduction a little better.

Never tried that--good idea. The Philips model has BF=16.9 for the
BFT25A vs. BF=85, fwiw.

A BFT25 is a better diode than practically any official diode.

Yup, an old fave for that. One of our products uses a small InGaAs
photodiode with a TIA whose feedback is 200M parallelled with two BFT25A
C-B junctions in series, with a feedback T-network that preserves the
full 5-V output range. It's for a more qualitative sort of application,
so we don't care much about the diode tempco. That combination gives
high sensitivity and constant gain in low light, and in brighter light
it preserves the the highest sensitivity consistent with not railing the
amp.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Phil Hobbs]

On 9/4/19 11:45 AM, Tauno Voipio wrote:

On 4.9.19 16:37, George Herold wrote:

Thanks that's interesting.  I made this 'audio' double balanced mixer
with
four diodes in a ring.  I found that diode connected transistors (B-C
shorted)
worked 'better' than regular signal diodes.

George H.

Half a century ago they were called transdiodes. The idea is that
the current gain of the transistor helps suppressing the effect
of the base resistance, and the transdiodes followed better the
theoretical diode equation.

It also greatly delays the onset of high-level injection conditions by
sucking the carriers out of the base ASAP.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Steve Wilson]

Piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 16:57, Steve Wilson wrote:
j.ponte@student.utwente.nl wrote:
I did a teardown of one and reverse-engineered it:
https://imgur.com/jRxgMy4

What did you tear down and where did you get it?

According to whit3rd, the AD 310 sat in a teflon bushed socket. Your
circuit would definitely not fit in a teflon socket.

(Still trying to make sense of this thing

Thanks

The teardown AD310 in the wonderful photos had been soldered into the
wider world judging by the tinned tips of the tall pins. However page 98
of the AD catalog at

https://ia801606.us.archive.org/35/items/AnalogDevicesDataAcquisitionPro
ductsCatalog1979/AnalogDevicesDataAcquisitionProducts1979_text.pdf

does describe a mating socket AC1017 that users could elect to use
instead of direct soldering. I guess the two input sockets of that
AC1017 would have teflon bushings. Teflon wouldnt have been necessary on
all the other i/o pins so it is possible those were insulated less
extravagantly. In the teardown photos one can clearly see the "hot" hi-Z
input circuit internally is on teflon standoffs so it makes sense to
keep up the high grade insulation.

piglet

The data for the AD310 and 311 starts on page 99. It shows socketed
versions of the ic's. Both ic's and the mating socket dimensions are shown
on page 100.

The circuit that j.ponte@student.utwente.nl showed does not correspond to
the information on either of these pages. The teardown was of some other
circuit. I am trying to find out what it was.

Thanks for the AD catalog link!

 

[piglet]

On 04/09/2019 6:58 pm, Steve Wilson wrote:

Piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 16:57, Steve Wilson wrote:
j.ponte@student.utwente.nl wrote:
I did a teardown of one and reverse-engineered it:
https://imgur.com/jRxgMy4

What did you tear down and where did you get it?

According to whit3rd, the AD 310 sat in a teflon bushed socket. Your
circuit would definitely not fit in a teflon socket.

(Still trying to make sense of this thing

Thanks


The teardown AD310 in the wonderful photos had been soldered into the
wider world judging by the tinned tips of the tall pins. However page 98
of the AD catalog at

https://ia801606.us.archive.org/35/items/AnalogDevicesDataAcquisitionPro
ductsCatalog1979/AnalogDevicesDataAcquisitionProducts1979_text.pdf

does describe a mating socket AC1017 that users could elect to use
instead of direct soldering. I guess the two input sockets of that
AC1017 would have teflon bushings. Teflon wouldnt have been necessary on
all the other i/o pins so it is possible those were insulated less
extravagantly. In the teardown photos one can clearly see the "hot" hi-Z
input circuit internally is on teflon standoffs so it makes sense to
keep up the high grade insulation.

piglet

The data for the AD310 and 311 starts on page 99. It shows socketed
versions of the ic's. Both ic's and the mating socket dimensions are shown
on page 100.

The circuit that j.ponte@student.utwente.nl showed does not correspond to
the information on either of these pages. The teardown was of some other
circuit. I am trying to find out what it was.

Thanks for the AD catalog link!

My bad, I gave you the page numbers as shown on the pages not the pdf
file numbering.

In what way does the teardown circuit not correspond to the datasheet?

piglet

 

[Steve Wilson]

piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 6:58 pm, Steve Wilson wrote:
Piglet <erichpwagner@hotmail.com> wrote:

The data for the AD310 and 311 starts on page 99. It shows socketed
versions of the ic's. Both ic's and the mating socket dimensions are
shown on page 100.

The circuit that j.ponte@student.utwente.nl showed does not correspond
to the information on either of these pages. The teardown was of some
other circuit. I am trying to find out what it was.

Thanks for the AD catalog link!

My bad, I gave you the page numbers as shown on the pages not the pdf
file numbering.

In what way does the teardown circuit not correspond to the datasheet?

piglet

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.

 

[George Herold]

On Thursday, September 5, 2019 at 8:47:06 AM UTC-4, piglet wrote:

On 04/09/2019 11:12 pm, Steve Wilson wrote:

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.




According to the catalog page dimensions the AC1017 socket pins are
pitched 0.2" (c.5mm) apart and the two rows are 2.2" (c.56mm) apart.

The teardown photos do not include a ruler but the dimensions of TO-92
transistors are well known, body height and diameter are both approx
0.2" (c.5mm). Using TO-92 as the yardstick on the photos I estimate the
pins pitch at 0.2" and rows spacing (11 x TO-92) at 2.2". It looks to me
very like the teardown device would nicely fit the AC1017 socket. The
pin functions on the teardown device agree with the 310 datasheet.

Do you have any other source for the AD310 circuit to say why the
teardown device is not a AD310? The patent is many years older than the
teardown device and the exemplar circuits in the patent might bear no
relation to any real world production. Fig 5 in the patent does look
rather like the teardown device bridge configuration. Please can you
share the reasons for doubting the teardown device is a 310?

piglet

There's a nice 'how it works' schematic of 310 in the catalog you linked to.
That mostly lines up with the hand drawn schematic above (by the student.)
Though the hand drawn one seems to be missing the synchronous detection
piece... done with diodes... I'm still trying to piece out how the whole
circuit works... Looking at it as a capacitance sensor makes some sense to me.
But I don't quite get the input bridge type circuit.

Fun stuff. (Steve gets grumpy sometimes.)

George H.

 

[piglet]

On 04/09/2019 11:12 pm, Steve Wilson wrote:

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.

According to the catalog page dimensions the AC1017 socket pins are
pitched 0.2" (c.5mm) apart and the two rows are 2.2" (c.56mm) apart.

The teardown photos do not include a ruler but the dimensions of TO-92
transistors are well known, body height and diameter are both approx
0.2" (c.5mm). Using TO-92 as the yardstick on the photos I estimate the
pins pitch at 0.2" and rows spacing (11 x TO-92) at 2.2". It looks to me
very like the teardown device would nicely fit the AC1017 socket. The
pin functions on the teardown device agree with the 310 datasheet.

Do you have any other source for the AD310 circuit to say why the
teardown device is not a AD310? The patent is many years older than the
teardown device and the exemplar circuits in the patent might bear no
relation to any real world production. Fig 5 in the patent does look
rather like the teardown device bridge configuration. Please can you
share the reasons for doubting the teardown device is a 310?

piglet

 

[George Herold]

On Thursday, September 5, 2019 at 12:16:56 PM UTC-4, j.p...@student.utwente.nl wrote:

On Thursday, September 5, 2019 at 12:12:55 AM UTC+2, Steve Wilson wrote:
piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 6:58 pm, Steve Wilson wrote:
Piglet <erichpwagner@hotmail.com> wrote:

The data for the AD310 and 311 starts on page 99. It shows socketed
versions of the ic's. Both ic's and the mating socket dimensions are
shown on page 100.

The circuit that j.ponte@student.utwente.nl showed does not correspond
to the information on either of these pages. The teardown was of some
other circuit. I am trying to find out what it was.

Thanks for the AD catalog link!


My bad, I gave you the page numbers as shown on the pages not the pdf
file numbering.

In what way does the teardown circuit not correspond to the datasheet?

piglet

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.

I'm not sure what makes you think it is not an AD310.
I got it on aliexpress of all places, the case has some wear and clearly came off something they salvaged. It was about 15$.
The bridge design also matches the one shown in the AD documents.

I could put the case up if you really want to see but there's nothing to it, just a metal shell.

The synchronous detection is done by the second JFET, to the right in the schematic, acting as a sort of chopper switch. The result of that operation is integrated.

Re 2nd Jfet: Ahh good! There were a lot of nodes labeled C or G and
it wasn't clear to me what went where.

GH

 

On Thursday, September 5, 2019 at 12:12:55 AM UTC+2, Steve Wilson wrote:

piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 6:58 pm, Steve Wilson wrote:
Piglet <erichpwagner@hotmail.com> wrote:

The data for the AD310 and 311 starts on page 99. It shows socketed
versions of the ic's. Both ic's and the mating socket dimensions are
shown on page 100.

The circuit that j.ponte@student.utwente.nl showed does not correspond
to the information on either of these pages. The teardown was of some
other circuit. I am trying to find out what it was.

Thanks for the AD catalog link!


My bad, I gave you the page numbers as shown on the pages not the pdf
file numbering.

In what way does the teardown circuit not correspond to the datasheet?

piglet

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.

I'm not sure what makes you think it is not an AD310.
I got it on aliexpress of all places, the case has some wear and clearly came off something they salvaged. It was about 15$.
The bridge design also matches the one shown in the AD documents.

I could put the case up if you really want to see but there's nothing to it, just a metal shell.

The synchronous detection is done by the second JFET, to the right in the schematic, acting as a sort of chopper switch. The result of that operation is integrated.

 

On Wednesday, September 4, 2019 at 9:27:05 AM UTC+2, piglet wrote:

On 03/09/2019 11:35 pm, j.ponte@student.utwente.nl wrote:

What do you think of the feedback to the JFET source? My best guess is that it's there to bootstrap the capacitance.


True, it will bootstrap capacitance somewhat but I think more simply it
is there to stabilize the open-loop gain.

piglet

Hmm, perhaps. Although I would expect it to be quite a bit speedier than the opamp-based integrator thingy, I doubt its associated poles show up above 0dB.

 

[Piglet]

On 05/09/2019 14:26, George Herold wrote:

On Thursday, September 5, 2019 at 8:47:06 AM UTC-4, piglet wrote:
On 04/09/2019 11:12 pm, Steve Wilson wrote:

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.




According to the catalog page dimensions the AC1017 socket pins are
pitched 0.2" (c.5mm) apart and the two rows are 2.2" (c.56mm) apart.

The teardown photos do not include a ruler but the dimensions of TO-92
transistors are well known, body height and diameter are both approx
0.2" (c.5mm). Using TO-92 as the yardstick on the photos I estimate the
pins pitch at 0.2" and rows spacing (11 x TO-92) at 2.2". It looks to me
very like the teardown device would nicely fit the AC1017 socket. The
pin functions on the teardown device agree with the 310 datasheet.

Do you have any other source for the AD310 circuit to say why the
teardown device is not a AD310? The patent is many years older than the
teardown device and the exemplar circuits in the patent might bear no
relation to any real world production. Fig 5 in the patent does look
rather like the teardown device bridge configuration. Please can you
share the reasons for doubting the teardown device is a 310?

piglet

There's a nice 'how it works' schematic of 310 in the catalog you linked to.
That mostly lines up with the hand drawn schematic above (by the student.)
Though the hand drawn one seems to be missing the synchronous detection
piece... done with diodes... I'm still trying to piece out how the whole
circuit works... Looking at it as a capacitance sensor makes some sense to me.
But I don't quite get the input bridge type circuit.

Fun stuff. (Steve gets grumpy sometimes.)

George H.

Hi George,

The synchronous detection in the teardown is a synchronous chopper
switch, pretty much the same as in the Philbrick P2 and chopper amps
going way back. Effectively works the same way as the commutating diodes
detector depicted in the principle of op sketch but requires more
filtering I think. I guess a one transistor switch is easier than
messing around with yet another transformer and matching diodes and so on.

I never tried to equate these varactor dc op-amps with RF parametric
amps. Instead I envisage them as a breed of chopper amp but with
(probably) lower conversion gain but the great advantage of zero dc load
current. So to backtrack, in a 1930-60s style chopper amp 1mV of dc
input gets chopped and makes 1mV of AC for easy amplification with old
technology. But the input gets loaded somehow. In a varactor op-amp with
zero volts input the two sides of the bridge are balanced and no AC
output exists. As the input unbalances the operating point shifts and
the two varactors capacitances unbalance and a small AC output exists.
Depending on the diode's V-C slopes the sensitivity could be lower -
e.g. 1mV of dc input may make less than 1mV of AC but hey gain is cheap
at AC. But because the diodes are never conducting the DC input bias
current is near zero. A very creative solution way back.

Bob Pease's story of the P2 is fun reading, especially how a quirk of
the board layout stymied competitors' attempts to copy.

Yes, fun stuff!

piglet

 

On Thursday, September 5, 2019 at 7:45:29 PM UTC+2, j.p...@student.utwente.nl wrote:

On Thursday, September 5, 2019 at 6:31:33 PM UTC+2, George Herold wrote:
On Thursday, September 5, 2019 at 12:16:56 PM UTC-4, j.p...@student.utwente.nl wrote:
On Thursday, September 5, 2019 at 12:12:55 AM UTC+2, Steve Wilson wrote:
piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 6:58 pm, Steve Wilson wrote:
Piglet <erichpwagner@hotmail.com> wrote:

The data for the AD310 and 311 starts on page 99. It shows socketed
versions of the ic's. Both ic's and the mating socket dimensions are
shown on page 100.

The circuit that j.ponte@student.utwente.nl showed does not correspond
to the information on either of these pages. The teardown was of some
other circuit. I am trying to find out what it was.

Thanks for the AD catalog link!


My bad, I gave you the page numbers as shown on the pages not the pdf
file numbering.

In what way does the teardown circuit not correspond to the datasheet?

piglet

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.

I'm not sure what makes you think it is not an AD310.
I got it on aliexpress of all places, the case has some wear and clearly came off something they salvaged. It was about 15$.
The bridge design also matches the one shown in the AD documents.

I could put the case up if you really want to see but there's nothing to it, just a metal shell.

The synchronous detection is done by the second JFET, to the right in the schematic, acting as a sort of chopper switch. The result of that operation is integrated.

Re 2nd Jfet: Ahh good! There were a lot of nodes labeled C or G and
it wasn't clear to me what went where.

GH

I finally decided to properly draw the schematic. It should be up at the end of the teardown imgur post.
Otherwise here's a direct link: https://imgur.com/QHB25jC

It shows up black for me (firefox) unless clicked for some reason.

If anyone wants the postscript file or sees errors, please let me know.

Oh the oscillator is pretty badly wrong. I shouldn't do these when I'm tired.
Maybe I'll fix it later.

 

On Thursday, September 5, 2019 at 6:31:33 PM UTC+2, George Herold wrote:

On Thursday, September 5, 2019 at 12:16:56 PM UTC-4, j.p...@student.utwente.nl wrote:
On Thursday, September 5, 2019 at 12:12:55 AM UTC+2, Steve Wilson wrote:
piglet <erichpwagner@hotmail.com> wrote:

On 04/09/2019 6:58 pm, Steve Wilson wrote:
Piglet <erichpwagner@hotmail.com> wrote:

The data for the AD310 and 311 starts on page 99. It shows socketed
versions of the ic's. Both ic's and the mating socket dimensions are
shown on page 100.

The circuit that j.ponte@student.utwente.nl showed does not correspond
to the information on either of these pages. The teardown was of some
other circuit. I am trying to find out what it was.

Thanks for the AD catalog link!


My bad, I gave you the page numbers as shown on the pages not the pdf
file numbering.

In what way does the teardown circuit not correspond to the datasheet?

piglet

The pcb shown in the teardown circuit is too big to fit on the AC1017
socket.

I haven't had time to study the patent, but I suspect the AD310-311
schematic will be competely different from the circuit shown in the hand-
drawn schematic.

The teardown circuit is completely different from the AD310-311. I am
trying to find out what it is from and where it came from.

I'm not sure what makes you think it is not an AD310.
I got it on aliexpress of all places, the case has some wear and clearly came off something they salvaged. It was about 15$.
The bridge design also matches the one shown in the AD documents.

I could put the case up if you really want to see but there's nothing to it, just a metal shell.

The synchronous detection is done by the second JFET, to the right in the schematic, acting as a sort of chopper switch. The result of that operation is integrated.

Re 2nd Jfet: Ahh good! There were a lot of nodes labeled C or G and
it wasn't clear to me what went where.

GH

I finally decided to properly draw the schematic. It should be up at the end of the teardown imgur post.
Otherwise here's a direct link: https://imgur.com/QHB25jC

It shows up black for me (firefox) unless clicked for some reason.

If anyone wants the postscript file or sees errors, please let me know.