MightyOhm.com

A few of suggestions for the next rev of the board...

Bring the power lines out to a header (pins on the existing serial header?) so that the board can be more easily powered by something other than the batteries. I've got a CP2012 based USB-UART adapter I'm going to be using with the GC kit to get the serial into the computer. It has 3.3V available and I plan to wire than in to power the board from the USB connection instead of batteries.

Alter the HV section to produce at least 1KV (and handle a heavier load) so that US geiger tubes and photomultipliers can be driven.

Add pads for a BNC connector (or a header) so that an already wired external probe can be connected.

Thanks!

I also support bringing the power supply rail out to the serial connector (for the same reason).

Actually, the other thing that I'd like would be to add three more resistors and a capacitor to make a cheapo ADC for measuring the HT. A 1G/2M ohm voltage divider on one of the AIN lines, and a simple RC on the other AIN, using one of the spare digital outputs to drive the R. This would allow a low fidelity reading of the tube voltage. [Actually, I just checked and 1GOhm resistors are stunningly expensive. In the 100MOhm range, they start to become affordable. What is the HT power supply capability (in mA)?]

http://www.atmel.com/dyn/resources/prod ... oc0942.pdf

Of course, a little extra code would be required as well.

Philip

Hi Philip!

Great suggestions, thanks!

Power is already available on the PULSE header. You could power the board by the PULSE header, but you will lose the functionality of the power switch. Pin 1 (with the arrow) is positive. For now I would avoid going over 3.3V, as I haven't done much testing at 5V (but it should work).

1kV is beyond the ability of the high voltage supply. There are some hard limits to the driver transistor breakdown voltage, this restricts the maximum voltage to around 600V. A different HVPS topology would address this, at added expense. Also, making a simple supply that can be adjusted from 300-1000V would be challenging!

The HV power supply current capability is on the order of microamps. It is extremely sensitive to loading, and any attempts to measure the voltage will no doubt load it down (which could be compensated, to a degree). There is a huge benefit to the "wimpy" high voltage supply - it is very safe to handle and not capable of delivering a significant shock. This is a very important feature of the kit.

It is worth noting that the input impedance of the AVR ADCs are on the order of 10Kohms, so feeding those inputs will require more than just a high impedance voltage divider. But yes, it is an interesting idea and the ability to measure the HV supply in software would be quite useful.

Again, thanks for your suggestions, and I encourage you to experiment and hack the kit - it is open source for a reason.

Here you go; this power supply circuit should operate safely up to 800V (it's capable of producing over 1500V, but doing so exceeds the component ratings). It takes Jeff's design and adds a split coil inductor and voltage doubler to generate a higher voltage while staying within the component rating limits, and a simple zener clamp for better voltage regulation. It has been tested and operates on 3V to 5V. It's not very energy efficient though, drawing 15 to 20 mA.

It'd be real nice if the MightyOhm kit included a 1G resistor, but they are available for less than $1 each at mouser.com (part #279-RGP0207CHK1G0). Following the instructions for the kit, the HV was well over 500V (yes, using fresh batteries), which exceeds the SBM-20 tube voltage rating. So calibrating it with a HV probe of some kind is a good idea.

500V is on the high side. The kits usually power up closer to 400V with VR1 adjusted mid-way, but component tolerances will cause some variation. The SBM-20 is an extremely forgiving tube and will operate significantly beyond the normal 350-475V operating voltage range without breaking down.

I would suggest to include a 3.3V regulator from pin 3 of the serial connector as an additional power source.

Why ?

The FTDI TTL-232R-3V3 passes the +5V USB power to pin 3 (red cable).
So you will be able to safely run the Geiger from USB power.

I have already patched my device (see pic)

More suggestions:

* Remove the 555 timer chip - replace its functionality with one of the ATtiny's timers. (reduce component count)
* Replace the current limiting circuit with software auto-adjust.
* [Replace MCU with ATmega32U2 - add USB programming, power and interface all-in-one]
* [add slim SMD design to fit everything (including 18650 LiIon battery and USB charger IC) into a hanheld tube or flashlight shell]

BR Peter

Hi Peter,

Thank you for your suggestions. Using an internal timer is a clever idea and one that I have not experimented with.