veenstra (comecer) VIK 202

[r6502]

Hello all,

time for a tear down of the Veenstra electronic of a VIK 202 ionisation chamber.

I got several Veenstra VIK 202 electronics and ionisation chambers. These kind of measurement devices are used in the nuclear medicine to determine the overall activity of a radioactive tracer, before it is applied to a patient. The tracers are used to get check if a patient has cancer because the cancer cells do consume a lot of energy. The used tracers use radioactive marked glucose, called FDG, to determine the location of the cancer.

At the weekend at 1st , I disassembled the electronics of the ionisation chambers, and also made some measurements  to check the performance of the electronics.

At 1st what is an ionisation chamber. You can check some explanation on Wikipedia:   
https://en.wikipedia.org/wiki/Ionization_chamber.
The chamber will produce a current, proportional to activity of the radioactive sample measured with the chamber. The current increases linear with the activity of the sample.

Due to the fact, that I do not have radioactive samples, I had to use a current source to check the response of the measurement electronics. Therefore I disconnected the internal battery and the connection to the ion chamber and used my Keithley 263 current source to inject a defined current into the preamp of the electronic to check its linearity.

I started with 1 pA and increased up to 1µA, but the device stopped at about 670nA, so it makes no different if I source 670nA or 700nA into the device, but until this current, the response was really linear, see below.

1pA in current:
>Channel1: 0.803, Channel2: -0.251, Activity: 0.000e+00
>Channel1: 0.799, Channel2: -0.251, Activity: 0.000e+00
>Channel1: 0.796, Channel2: -0.251, Activity: 0.000e+00
5pA in current:
>Channel1: 4.741, Channel2: -0.251, Activity: 0.000e+00
>Channel1: 4.745, Channel2: -0.251, Activity: 0.000e+00
>Channel1: 4.745, Channel2: -0.251, Activity: 0.000e+00
10pA in current:
>Channel1: 9.730, Channel2: -0.251, Activity: 0.000e+00
>Channel1: 9.735, Channel2: -0.251, Activity: 0.000e+00
>Channel1: 9.732, Channel2: -0.251, Activity: 0.000e+00
50pA in current:
>Channel1: 49.459, Channel2: -0.250, Activity: 0.000e+00
>Channel1: 49.456, Channel2: -0.250, Activity: 0.000e+00
>Channel1: 49.461, Channel2: -0.250, Activity: 0.000e+00
500pA in current:
>Channel1: 496.280, Channel2: -0.245, Activity: 0.000e+00
>Channel1: 496.295, Channel2: -0.245, Activity: 0.000e+00
>Channel1: 496.293, Channel2: -0.245, Activity: 0.000e+00
50nA in Current:
>Channel1: 49709.563, Channel2: -19.975, Activity: 0.000e+00
>Channel1: 49709.582, Channel2: -19.974, Activity: 0.000e+00
>Channel1: 49709.586, Channel2: -19.974, Activity: 0.000e+00
500nA input current:
>Channel1: 499607.656, Channel2: -126.646, Activity: 0.000e+00
>Channel1: 499607.594, Channel2: -126.645, Activity: 0.000e+00
>Channel1: 499607.688, Channel2: -126.641, Activity: 0.000e+00

The current was  sourced into channel 1 of the instrument. It was sourced direct into the PCB, without using the coax cable, because that was generating too much noise and I got less stable readings.

The dChip used for measuring has 2 channels for current monitoring, pin 1 used for Channel 1 and pin 28 used for channel 2. In this application only channel 1 was used.

Has anybody seen an integrated current monitor with this input range and its linearity over several decades of input current? On the used chip, U7(I marketed the chip red in the photo, where the PCB is visible), on the PCB the type information of the chip was removed by Veenstra, so I think, it is no custom chip. Does anybody have an idea what chip Veenstra could have used here it has 28 pins and  2 inputs on pins 1 and 28? It would be very interesting for me.

Please have a look on the photos of the device. The last 2 Photos are from the ionisation chamber with included electronics. The measuring chamber was shielded with some lead. The lead was covered with some plastic tape because of the toxic lead.

Kind regards - Guido

edit: spelling ...

[doktor pyta]

For the first time I see 50 lithium coin batteries arranged to create >150V battery.
I always try to understand why do engineers went certain path while designing.
Here we have worth 20 EUR ultra low noise, 'floating', high reliability, stable bias generator that will last for 10 years and at low engineering cost -sounds legit.

[Kleinstein]

Given the large current range, the special 28 pin chip would likely be a kind of integrator.  TI has some such integrators (e.g. IVC102) that can be used to measure small currents.
My guess for the chip would be a DDC112, a  ADC with current input on pins 1 and 2. 

[r6502]

@Kleinstein
Perfect - the DDC112 is the chip I was searching for. I think Venstra used really the BB reference design, also with the reference voltage. Thy also used the OPA350 for the reference voltage filtering and buffering.

I did not had in focus, that BB (now TI) build ADCs with direct charge input usable for this kind of applications. In our company we did some tests with the good old IVC102, but decided to use the ADA4530 instead with some additional external components.

Yesterday I started to do some measurements with the scope on the lines connected to the µC, to get more details,  but was not finished. The measurements will fit exact to what BB describes in the data sheet of the chip.

Thanks!

@ doktor pyta
yes, with the 50 lithium coins it is really nice design for an low noise bias generator - and it is working well.-

Very interesting is also how they measure the voltage of the bias battery: the reed contact will be closed and through the 10G resistor a huge   current of about 16nA (when battery is fine) is generated, proportional to the voltage of the battery ...

Kind regards - Guido

[BD1QMP]

DDC112 was used in CT scanner DAS board. the second hand chip is very cheap. only half a US$