chance pi build

[baum7154]

Fascinating stuff, thanks for the info!

[baum7154]

A little more detail on the other A-D samples I saw. In the short pulse portion of the TX I think there were 7 groups of two samples evenly spread within the 15 short pulses. They came in pairs, each sample 12.5 wide, a 1us break and another 12.5 us sample for a total of 14 samples. I'll see if I can get a good picture and post it this evening.

Dan

[baum7154]

A little more detail on the other A-D samples I saw. In the short pulse portion of the TX I think there were 7 groups of two samples evenly spread within the 15 short pulses. They came in pairs, each sample 12.5 wide, a 1us break and another 12.5 us sample for a total of 14 samples. I'll see if I can get a good picture and post it this evening.

Dan

CORRECTION!!

Was looking with the scope last night and counted the short pulses again. There are sixteen (16) short pulses in the string, not 15 as I said previously. Sorry for the misinfo.

Setting critical damping on the scope while listening to the audio from the CHANCE PI shows that at the precise point where critical damping is met, the audio dramatically changes from a raspy buzz to melodical response. Pushing ZERO (RESET) at this point quiets it down and the detector damping is optimized. If you want to slightly overdamp you can add a small tweak to the pot. I will be adding a Variable Damping option to this detector assembly so that it can be optimized for different soils using the audio response in the field. No scope needed from what I have seen but will test in the field and post results here. This feature is also useful when changing from one coil to another.

Dan

[baum7154]

CORRECTION!!

Was looking with the scope last night and counted the short pulses again. There are sixteen (16) short pulses in the string, not 15 as I said previously. Sorry for the misinfo.

Setting critical damping on the scope while listening to the audio from the CHANCE PI shows that at the precise point where critical damping is met, the audio dramatically changes from a raspy buzz to melodical response. Pushing ZERO (RESET) at this point quiets it down and the detector damping is optimized. If you want to slightly overdamp you can add a small tweak to the pot. I will be adding a Variable Damping option to this detector assembly so that it can be optimized for different soils using the audio response in the field. No scope needed from what I have seen but will test in the field and post results here. This feature is also useful when changing from one coil to another.

Dan

I should note that the detector Guard Interval was set to it's lowest limit of '10' for this test/adjustment. Setting it higher may not work as well because GI=10 gives the fastest sample available. The particular coil used in this test crosses zero at 6us after the end of the 120us long pulse. The A-D sample of this period begins 8us after the end of the long pulse and lasts for 12.5us. Slower coils may require higher setting of the GI before damping adjustment using the audio can be accomplished.

Dan

[green]

I should note that the detector Guard Interval was set to it's lowest limit of '10' for this test/adjustment. Setting it higher may not work as well because GI=10 gives the fastest sample available. The particular coil used in this test crosses zero at 6us after the end of the 120us long pulse. The A-D sample of this period begins 8us after the end of the long pulse and lasts for 12.5us. Slower coils may require higher setting of the GI before damping adjustment using the audio can be accomplished.

Dan

I think the A-D samples the input when pin 5 first goes low. The 12.5 usec is the time required to read the sample. Have you looked at amplifier out and pin 5 while changing GI, damping resistor, etc. to see what amp. out looks like when sampling?

[baum7154]

I probably have but don't recall the exact details/timing/responses of the output. Will try to look at it tonight. Seems I do recall the decay took longer maybe due to amp saturation. I'll let you know.

Dan

[baum7154]

I think the A-D samples the input when pin 5 first goes low. The 12.5 usec is the time required to read the sample. Have you looked at amplifier out and pin 5 while changing GI, damping resistor, etc. to see what amp. out looks like when sampling?

The A-D sample window stays open for 12.5us after pin 5 goes negative. I think this is to allow time for signals occurring within that time to be captured. Maybe that is incorrect.

[baum7154]

I think the A-D samples the input when pin 5 first goes low. The 12.5 usec is the time required to read the sample. Have you looked at amplifier out and pin 5 while changing GI, damping resistor, etc. to see what amp. out looks like when sampling?

Attached here are some pictures with a description of each screen shot:

1. This picture is of the coil flyback measured at the diodes on the amp input. The horizontal line at midpoint is the A-D sample window


2. This picture is of the last of 8 short pulses and the 120us long pulse and coil flyback,again taken at the diodes. The Guard Interval moves the single A-D sample shown at the far right 0.9us for each increment of GI input. All others are unaffected by GI settings.


3. This picture is of the 16 short pulses with their corresponding A-D dual samples in the 2nd horizontal row from the top.


4. This picture was taken at the amp output and at pin 5 of the A-D. It shows the A-D sample at its position with a Guard Interval of 10. This shot is of the improperly damped coil and the second flyback image at the right of the trace is there because damping is not correct. The detector audio is a raspy buzz in this state . Note the increased flyback time at the amp output compared to the measurement at the input to the amp.


5. This picture is taken at the amp out and A-D pin 5. It shows a properly damped coil and the A-D sample. Note that the image to the right from the previous shot is now gone. The detector audio is now in a quiet state.

[green]

Thanks for the photos. Looking at jpg no. 5 the sample goes low while amplifier out is saturated. I must be wrong about how the A-D works.

[Tepco]

Actually this can work, but just barely. Chip acquires input signal during first 1.5 cycle of serial clock signal when enabled, or some 1.35uS. During this time, internal sample&hold circuit is “charged” normally expecting steady voltage, but due to rapidly changing waveform, voltage is integrated and ADC can produce meaningful output, despite one part of waveform is in saturation. Very critical “on the edge” adjustment, fine to determine damping resistor value, but from this point maybe better to increase GI few steps to sample on flatter waveform portion and check detector behavior. Not sure how well OP37 can cope with this, part of decay may be limited with amplifier saturation speed, two stage amp can be considered here. I'm finishing my hardware to make some tests, will post on this later.

[baum7154]

Yes I have been thinking about how I can integrate a dual opamp neatly in place on this detector. Clearly a dual amp would benefit performance.

Also when detecting it is better to set GI 1 or 2 increments out to get A-D just outside of the flyback curve.

[Tepco]

Another chance for Chance. I managed to find few years old readily made PCB (heavily oxidized but fixed after some careful cleaning), single sided version, and almost all parts in my junkbox , and here it is: I hope during weekend everything will be completed and programmed, ready for testing. At this point not everything is “original”, I don't have OP37 at hand, have to order , transistors are IRF 9630, not 9640, fast diodes are UF1007, and this version is without new hardware modification, I will add this later easily to see what difference it makes. This detector never cached up, but probably we underestimated it's ability, with some tweaking, mostly coil oriented, this can be really nice DIY project. Something capable for sub-10uS sampling, multiperiod timing and effective ground balance, at least some form of discrimination etc. hand made is always interesting. At least I will properly document circuit operation. Also I want to try some tricks for fine tuning VDI response, idea appeared some time ago on some foreign (probably not even Russian) site, but too late to try it, at a time I already disassembled my hardware.

[baum7154]

Good Luck TEPCO! I am anxious to have others collaborating on optimization of this worthy project. Right now my CHANCE does detect small gold pretty well but we all know it can be even better with some serious focus on things like coils, amps, damping etc. I look forward to hearing about your build and results.

Dan

[Waikiki_Sweep]

Another chance for Chance. I managed to find few years old readily made PCB (heavily oxidized but fixed after some careful cleaning), single sided version, and almost all parts in my junkbox , and here it is: I hope during weekend everything will be completed and programmed, ready for testing. At this point not everything is “original”, I don't have OP37 at hand, have to order , transistors are IRF 9630, not 9640, fast diodes are UF1007, and this version is without new hardware modification, I will add this later easily to see what difference it makes. This detector never cached up, but probably we underestimated it's ability, with some tweaking, mostly coil oriented, this can be really nice DIY project. Something capable for sub-10uS sampling, multiperiod timing and effective ground balance, at least some form of discrimination etc. hand made is always interesting. At least I will properly document circuit operation. Also I want to try some tricks for fine tuning VDI response, idea appeared some time ago on some foreign (probably not even Russian) site, but too late to try it, at a time I already disassembled my hardware.

As I can see from picture it is Chance layout from DesAlex. It is smaller, one side, easy to make PCB. I have etched one.
Only my problem was that lay file was printed a couple % smaller so 28 pin controller hard to fit in it.
Probably Sprint or my laser printer were wrong. Better check dimensions before etching.

You can download all files including "Chance layout from DesAlex.lay"
http://forum.cxem.net/index.php?app=...ttach_id=75463

Comments on Russian on forum:
http://forum.cxem.net/index.php?show...dpost&p=510118



layout.zip









Readme.txt

[Tepco]

This is not exactly DesAlex board but same design, I made this in Protel from scanned hand drawn board layout (Сборочный черт&#11.jpg ) time ago, in 2009 or 2010 I think. Few boards are made then, and few detectors built, earlier version software, but performance was not satisfactory. Later new software updates came, but by then I disassembled complete detector and tried new versions using just PCB, later cannibalized for parts. This one leftover PCB remained, design is known to work, will be nice platform for some tests.



Issue with resized PCB happened to me more than once, print from .pdf, or wrong paper format, letter instead of A4 or... Always noticed when PCB is etched, tin plated and drilled.