Universal PI Micro

[bugwhiskers]

UPIM update

Hi Ray,

I have tried to make at home a PCB of the circuit board posted earlier and it is just too tight, small tracks, no meat on the pads so I have decided to make the board larger. This will also make it easier for folks to make their own from the artwork. I have also decided to use a 40 pin DIP chip that will again make life a lot easier for the homebrewers.

With regards to the other project, the following is a brief outline:

The front end preamp, coil drive and damping resistor are the same as in most PI circuits, but thats where the similarity ends.

The circuitry following the preamp gets 64 samples (via switched capacitors) of the early part of the decay waveform. The time between each sample can be as low as 100nS. At that speed it is possible to capture 64 samples of the first 6.4 uS of the decay waveform. The capacitors are then switched in turn to the AtoD of the micro and the digitized result is stored in RAM.

One area of RAM holds an average of all the previous samples with user adjustable software lag and another area holds the most recent. The lag is necessary to prevent small changes being swamped by the averaging. Ground signal and preamp drift will become part of the average set of samples whereas any abrupt change will be seen as a target. Because the sampling is done very early in the decay it should be possible to discriminate.

What is unique about this approach is the the samples are gathered very quickly and then later processed at leisure between the TX pulses. This technique obviates the need for a super fast micro and a super fast AtoD.

The 10uS per division CRO shot shows, in the bottom trace and to the left, the 64 samples being taken (6.4 uS). The rest of the trace shows the time taken to digitize them and store to RAM.

regards
bugwhiskers

[johnno]

Hi BW is this an addon to your previous published cct?
or a new homebrew?
johnno

[Ray-NM]

BW
Looks like a very efficent way to get the data. That is the problem I am having with the Basic Stamp it is too slow to process the signal immediatly.
I will think on your approach and see if I can implement it your way.
Great Idea.
RayNM

[bugwhiskers]

Hi BW is this an addon to your previous published cct?
or a new homebrew?
johnno

Hi Johnno,

It's very new. The preamp stage is much the same as other PI's with the exception of a fast first stage and a fast high current drive for the second stage to charge/discharge the caps. Everything downstream is different.

regards
bugwhiskers

[bugwhiskers]

Screen option

The attached pic shows an example layout for the 64 S&H PI / UPIM.

The LCD is 128*64 dots.

The top left box shows the 64 samples (actually a couple less due to the box lines) The top right box is a zoom window of 8 of the samples of the left box.
The bar graph in the middle is signal strength and the 6 digits are user adjustable variables.

I am waiting on a couple of 240*128 dot screens to arrive, these will allow double the resolution and allow more room for other "things"

regards

bugwhiskers

[Ray-NM]

I like it.
Raynm

[Ray-NM]

do you have any code available for us to look at yet? My assembly language skills are quite rusty so I might have to look at it for a few months to understand how and what you are doing in the controller.
thanks
RayNM

[gef12]

like it

Hi BW
Yep it be looking good thus far , Waiting in anticipation

Gef fm OZ

[bugwhiskers]

64 S&H OVERVIEW

The attached drawing shows the 64 sample and hold concept.

After the coil pulse is terminated the micro switches the output of the preamp to each of the 64 caps in turn. The user has control of the width of each sample the smallest being 50nS. Straight after the 4053 is toggled over and the stored charge on each of the caps is fed to the micro's AtoD, digitised and stored to the micro's RAM.

There will be 2 sets of samples in RAM, the current and an averaged set of previous samples. A user adjustable variable called "LAG" will control how often the current set will be averaged with the previous set. This will allow changes to be seen before they get swamped by the averaging.

With total control over coil pulse width, start of sampling time and the area of the decay curve to be sampled should allow the results to be analysed with a view to discrimination. Experiments with a prototype have shown the voltage on the caps to be very stable and relatively immune to the noise generated by the micro.

The top left box of the display will show the difference between the current sample and the averaged previous samples. If the user sees an area of difference then the right top screen can be adjusted to show a zoomed view (8 samples wide) of that area. The signal strength can be linked to that area also. I have seen graphs of decay curves of different metals and the area of the most difference is in the knee area of the curve.

More to come soon.

regards
bugwhiskers

[bugwhiskers]

do you have any code available for us to look at yet? My assembly language skills are quite rusty so I might have to look at it for a few months to understand how and what you are doing in the controller.
thanks
RayNM

Hi Ray,

All my work thus far has been in modular format. Hardware to do this and Software to do that. I am getting close to getting it all together.

One of the first things you should do is get a screen of some sort running off your micro to show internal variables etc... makes it much easier to debug your code.

I recently won two 240*128 LCD screens on ebay for $12US each (plus freight). Do an ebay search on LCD graphic display.
I have code for driving any graphic screen that uses a Toshiba T6963C chip
or the Hitachi HD44780 chip.
The code I have works but needs to be commented very well so people know what each part does.

regards

bugwhiskers

[bugwhiskers]

Some source code for Ray-NM

Hi Ray,

Attached is a text file that is the source code for the program that is driving the LCD screen shown a few posts back. It's not fully commented but all the sub-routines have descriptive names like ss_bar = signal strength bar.

The commands within the LCD controller chip are too slow to execute so I have used a bit map approach. The micro holds in RAM an image of the screen. Clearing the image in RAM and setting pixels etc is performed much faster within the CPU. When the image is complete the whole lot is sent out to the LCD.

regards
bugwhiskers

[Ray-NM]

Thanks for the info I will read thru it.
Raynm

[bugwhiskers]

Circuit for comments please.

The attached .PDF shows a circuit for generating the voltage required by the Opamps and Micro above the battery voltage.

Things to note, PIN 3 of IC14 is NOT at GND potential.
The switch needs to be a double pole double throw with center off.

The operating sequence is as follows.

The switch is in the middle position initially, everything off.
The operator momentarily moves the switch to the down position connecting the battery to C101 (~4700uF). This powers IC14. The pullup resistor R4 causes Pin 5 of IC14 to be LOW, this allows C66(~100uF) to charge via D1 and D3.
The operator then moves the switch to the up position. This connects the negative battery terminal to the micros Gnd and the battery positive pin to D4. C67 will now have approximately 2 times the battery volatge at its positive pin. IC1 will now be functional and provide power to the micro.
The micro's Pin 5 is a PWM output that will start sending pulses to the Opto OK1 and on to IC14.
The operator then moves the switch back to the down position for normal operation.
Most circuits use a 7660 chip or similar but the available current won't power the 11 chips downstream and the display. The only other alternative was to use a second battery but that would be awkward.

Any comments on the circuit would be appreciated.


regards
bugwhiskers

[bugwhiskers]

Attachment ?

It really helps if I attach the file !

regards

bugwhiskers

[Stefan]

Power supply for BW's design

Hi Bugwiskers,

Been following your design with interest. I am using the Rabbit Core RCM4100 microprocessor as the basis of my PI design. The design is based on Candy’s patents so the sampling and pulse generation is somewhat complex. I was faced with the same problem you are having namely power supplies. Originally I decided to design and build my own power supplies using switch regulator chips. The problem with switch regulators is that high frequencies generated are feed back into the main battery supplies so carefull attention must be given to attenuating these signals.

However this was becoming time consuming so I looked at another means. I settled on commercially available DC/DC Converter modules made by TRACO Power. There is quite a range to choose from. The modules are fully shielded and have input filters for attenuation of the high frequencies. These modules are available from RS Components or Farnell but are somewhat expensive. If you consider the time spent on design, etc they are not that expensive and will save a lot on design.


Regards,

Stefan