Voodoo Main Discussion

[Qiaozhi]

Found some reject metal boxes that require the board to be a little wider and slightly shorter has anyone got a pcb file from there gerber? Secondly to drive the lcd via 16 pin ribbon is a nightmare It would be more logical to use IIC/I2C/TWI/SPI as in the Arduino Has anyone done software update to do this?

It would be less hassle to purchase an enclosure of the correct size.
Driving the LCD with a parallel interface provides the fastest connection. Using one of the serial interfaces may be too slow, and would also require a different LCD.

[GeoMax]

Hi Phred01,

to make a PCB wider is no problem... You can make it maybe 1 meter wider without problems. But to make it smaller it might be a problem. Maybe to make it 2..3 mm smaller it is possible. But not really much more. To get the PCB size you want you need probably completly redesign the whole PCB new.

[phred01]

Found some reject metal boxes that require the board to be a little wider and slightly shorter has anyone got a pcb file from there gerber? Secondly to drive the lcd via 16 pin ribbon is a nightmare It would be more logical to use IIC/I2C/TWI/SPI as in the Arduino Has anyone done software update to do this?

[Qiaozhi]

Since I am just starting out haven't organized a compiler yet any change for a copy of the hex file please

I think the hex file in the Voodoo.XC3.00.X.zip under dist/default/production.

[phred01]

Hi all, as mentioned in a previous post here's an updated software for the Voodoo should anyone be curious. Details are in the previous post. I've included some scope screenshot showing pulse and hybrid modes.

Yellow trace is Tx pulse at the junction of the anode of the diode and 470 ohm dummy load resistor.

Cyan trace is the main sample pulse at TP10.

Magenta is the EFE sample pulse at TP6.

Blue is the Disc sample pulse at TP11.

I've also included a thermal image of the board with dummy load. No surprises that the 75L09 regulator is getting pretty hot with almost 100mA total current draw from my PSU.

You can see the alternate disc pulses appearing in hybrid mode from the scope pics.

Since I am just starting out haven't organized a compiler yet any change for a copy of the hex file please

[Gunghouk]

Thanks George, I'm really pleased with the resultant super stable pulse trains and faster loop time in the software. Hopefully someone with a working detector can give some feedback on the performance vs 2.02. Sampling should be super smooth too.

[Qiaozhi]

Looks good.

[Gunghouk]

Hi all, as mentioned in a previous post here's an updated software for the Voodoo should anyone be curious. Details are in the previous post. I've included some scope screenshot showing pulse and hybrid modes.

Yellow trace is Tx pulse at the junction of the anode of the diode and 470 ohm dummy load resistor.

Cyan trace is the main sample pulse at TP10.

Magenta is the EFE sample pulse at TP6.

Blue is the Disc sample pulse at TP11.

I've also included a thermal image of the board with dummy load. No surprises that the 75L09 regulator is getting pretty hot with almost 100mA total current draw from my PSU.

You can see the alternate disc pulses appearing in hybrid mode from the scope pics.

[Qiaozhi]

Getting all the bits together On page 40 of the book there is a pix of the display with buttons a housing on a goose neck were can buy this or is there a STL file to print it up?

The display housing was from a White's MXT, but there is a 3D printed panel below the display to hold the buttons.
I've attached the FreeCAD design file, the STL file, plus the gcode and gco files for the switch panel.
However, I suspect you're really asking for the STL file for the complete housing. That (of course) does not exist, as it's a housing from White's.


Switch_Panel.zip

[Gunghouk]

Voodoo is 100% George.



This is generally true. In the case of a Voodoo, the ADC is sampling relatively slow signals so I doubt the sampling jitter will make a big difference. I don't recall the processing loop rate of Voodoo so I can't say for sure.

My personal preference is to run the ADC at a specific sample rate (say, 5ms) and use the data ready to trigger an interrupt, whereby this sets the loop rate of the software. But then, I also run the timers for the TX & demods autonomously. Voodoo uses a low-end PIC micro so it is pretty limited for doing stuff like this.

Hi Carl, I have been able to utilize the limited capability of the pic and eliminate(?) jitter while increasing overall speed and averaging accuracy as previously outlined.
I intend to make available soon a new version XC3.00 with these improvements, for testing by anyone interested with a functioning Voodoo detector.

All changes very verified on a Rigol
DHO 814, four channel oscilloscope.

The change details are :

***
XC3.0 (Neil Jolliff, aka Gunghouk)

DSP routine call moved to interrupt service routine to improve signal averaging, reduce main loop time and
Improve loop time consistency thereby reducing jitter.
DSP only runs every 1mS Timer1 call to ISR not on pulse timer0 ISR calls.

NOTE: This change enures consistent pulse timing regardless of any future code changes affecting main loop
execution time eg. UI mods.
Will probably improve detector response times and allow reduction in averaging sizes.

PI sampled on high sync pulse, Disc sampled on low sync pulse regardless of mode. This helps with DSP code speed
consistancy.

Timer value calculations simplified to ease understanding.
Disc pulse delay, after TX pulse, offset value added.

All timer offsets changed to suit above modifications.

Timing routine rationalised to reduce ISR run time helping to eliminate jitter.

Battery averaging changed to 1024 samples due main loop speed increase affecting battery display jitter.

Slight increases in LCD data delay times in order to eliminate occasional display artifacts occurring on 4x20
display.

PI oscillator frequency in common.h changed to 20Mhz

***
Future mods(?)
Single rotary encoder control instead of 4 buttons.
4x20 lcd or graphic display

***

Why bother you might ask.....

1. Because I am an engineer (albeit recently retired) and have nothing but time on my hands.
2. To keep learning new things (PICs & programming) and to refresh long ago forgotten or barely used skills (C programming)
3. Because I can and love doing it.
4. Fear of mushy retirement brain syndrome.

Regards to all

Neil

[phred01]

Getting all the bits together On page 40 of the book there is a pix of the display with buttons a housing on a goose neck were can buy this or is there a STL file to print it up?

[Qiaozhi]

Other than fragmented circuits in the book is there a complete schematic on one sheet

There is no complete "one sheet" schematic available.

[phred01]

Other than fragmented circuits in the book is there a complete schematic on one sheet

[kt315]

I'm also in the UK, and I have a couple of spare blank Voodoo REV-C PCBs. One of them has a 40-pin socket for the PIC soldered in place.
Send me a PM.

does this PIC have a SMD, 'bug', hull?

[phred01]

Thanx for the offer Have ordered boards from JLCPCB very easy to order & @ a very reasonable cost incl shipping 5 boards will be more than enough to build some detectors Will post my experience and modifications that I plan