Archaeology Resistivity Meter

Interesting ideas so far.  I meant to chime in earlier, but things have been pretty busy for me lately (building a deck and entertaining my Grandchildren again).

Several years ago, I was asked to sit in with some friends of my sister that work at a geotech company.  They were looking to build impedance measurement devices for soil surveys.  I came across this amazing looking chip from Analog Devices that looked like a great way to measure impedance (including a complex component).

Here is a snippet from the specification:

The AD5934 is a high precision impedance converter system solution that combines an on-board frequency generator with a 12-bit, 250 kSPS, analog-to-digital converter (ADC). The frequency generator allows an external complex impedance to be excited with a known frequency. The response signal from the impedance is sampled by the on-board ADC and a discrete Fourier transform (DFT) is processed by an on-board DSP engine. The DFT algorithm returns a real (R) and imaginary (I) data-word at each output frequency.Once calibrated, the magnitude of the impedance and relative phase of the impedance at each frequency point along the sweep is easily calculated using the following two equations:Magnitude = 22IR+Phase = tan−1(I/R) A similar device, available from Analog Devices, Inc., is the AD5933, which is a 2.7 V to 5.5 V, 1 MSPS, 12-bit impedance converter, with an internal temperature sensor, available in a  16-lead SSOP.

This might be of some help in your planning.

Good luck and let me know if you need any help on this project.

Hi Gene,

It's a super-interesting chip, I was keen to use it a few years ago for plant soil purposes, and for hydroponics - to try to see if the soil or liquid has nutrients. The idea being to have a signature of known good soil or water by sweeping through the spectrum. I never got to try it though sadly, the project moved on to something else.

It was felt that it could have had a lot of merit because then you could publish the signature, so others could try to replicate a yield (it wasn't going to be for farmers, more for home use), and to not waste nutrient. But, I have no idea in practice if the result would have been usable, or too inconsistent/variable.

I wish I'd done some work on it at the time, since it could have been useful for other purposes too.

The proposed design so far is one half of the impedance measuring system, but with digital processing. In theory it could be converted to an impedance measuring system with no additional hardware change, just a software upgrade, since the frequency will be know, and there will be some sync pulse from the FPGA, we just need to internally multiply with a 90 degree out of phase signal from that sync pulse too.

Hi a full graphics display wouldn’t be entirely necessary. It was done a few years back on a unit by T R SYSTEMS. They seem to have disappeared though. It was a box of electronics mounted on the side of the frame and an android tablet mounted facing the operator. The connection was remote and I guess Bluetooth. It’s complicated when surveying to work out exactly where you are on a grid, what with 20 rows to survey and 400 readings to take. Seeing numbers relating to your position is good, but it can still be confusing. We also have the added complication of obstructions, such as trees where we can take a reading. A dummy reading is taken in these instances. I’ve posted up my build of the original 2003 resistivity logger and there is also the GEOSCAN RM85 offering, both I feel overcomplicated in the number of buttons and very small displays. Equally the RM FROBISHER unit has a minimalist design, but a nightmare menu system behind it.

Here's my promised block diagram of how I see the story so far:

The bit you can't read on the right says that it could be a different display with an SPI interface.

The Bridgetek thing I have in mind is an ME813A-WH50C (Farnell 2674200) but it is expensive - slightly cheaper from RS and similar things much cheaper from Mouser.

I've used the Microchip BT module and I know it works, an ESP part might be preferred by some because it could do WiFi as well.

I've put an SD card and USB OTG as options - they will need loads of software and probably a processor with more pins (than the 100 I've suggested).

If we want to do much processing and have a big display on the board it might be better to go for a faster processor but it will use more power.

I've suggested a low power version here.

From comments so far it may be that a 4 line LCD character display would do for starters - it has the virtue of cheapness and simplicity.

With care we could keep the interface dual purpose.

I'm thinking of at least 8 Mbytes of directly connected and soldered in flash.

Nothing is set in stone, additional scribbles and crossings out are welcomed.

The basic idea of this design is that it uses sine wave excitation (lots of existing stuff uses square wave which is horrible).

The phase sensitive detector is entirely in software.

The FPGA is tiny (ICE40UP5) and costs about £5 - it is for glue purposes rather than computation.

The proposed operator controls will be buttons and one or two rotary encoder type knobs.

These go through the FPGA to avoid the use of high speed asynchronous interrupts that rotaries require (the FPGA is there so why not use it ?)

Everything is software controlled - there are no hardware switches affecting filters, ranges, signal routing or stuff like that.

The expansion interface could work a multiplexer (sometime far in the future).

MK

This is great.  If shabaz hasn’t already started I can create a PowerPoint version this evening.

That would be good but I was rather hoping that there would be some critical feedback, so it might be worth waiting a day or two so you will only need to do it once.

(Although a Power point version should be easier to mod than mine.)

MK

Hi Michael,

Thanks for this! The diagram is easy to follow. It all makes sense, it's great that there are options to go from character display to an SPI one too. Incidentally there's low-cost and quite large RA8875 displays that could be around for a while (they've been around for several years), I've used one and it has an early 2000 laptop feel to them, i.e. average quality, not bright when viewed from the side etc). They are SPI based too, so with no additional hardware work it could be possible to connect the higher-end BridgeTek panel, or the cheaper one, if one day someone extends the firmware for that. Here's an example of that: https://www.buydisplay.com/5-inch-tft-lcd-module-800x480-display-controller-i2c-serial-spi

Also it's very neat that the Bluetooth module capability and USB UART is there. I'm going to try connecting a phone or tablet to USB UART today, to see if it works, but I don't have a FT230 device to test unfortunately (and will likely need some different Android source). Anyway I believe the phone/tablet capability is an accoutrement and not totally essential, since there's a wealth of display and connectivity options in this design.

I just had one comment, regarding the DAC output, I was wondering if it could be nice handing that off to the FPGA too, to do the DDS, and have external DAC? I know it's not essential, but it could make the microcontroller code easier for anyone to extend it in future, since then there's no SPI writes on timer interrupts once it is set up, and the microcontroller can just read from SPI each time there's an interrupt from the FPGA, allowing developers to just concentrate on processing the results via DSP and then storing or displaying the data and UI controls. It also means that entire microcontroller portion could be replaced in future with (say) Pi for future non-archaeology uses too, e.g. as a lab instrument, provided it is fast enough to respond to an interrupt to read the data from the ADC via FPGA.

fmilburn it's great if you can get this into a ppt, but will think about the design more today and might have a few more comments later this evening.

I was going to draw the block diagram up this evening on cad.

Hi kltm,

I will leave the block diagram to you then...  Can stick a placeholder in there for the power as well.

Frank

I'd wondered about using an off chip DAC - it doesn't need to be fast -  1kHz update rate would do, 8kHz would be really nice. The processor has 12 bit DACs built in and can refresh via DMA so th processing burden is very light.

On the other hand if the FPGA does it then synchronising is easier (the uP can run at whatever speed it likes) .

I'll check out what DACs are cheap and easy to get.

Those displays look nice and cheap but the processor has to work hard if you use graphics. The cute thing about the Bridgetek controller is how much easier they make doing graphics and how little data you need to transfer over SPI.

MK

TI do a nice DAC available in 16 or 12 bit versions fro £4.61 or £2.03 (DAC80501 or DAC60501).

SPI in and voltage out. It has a built in voltage reference which we can use for everything else as well.

So we need to connect the DAC to the FPGA in the block diagram rather than to the micro.

MK

115200 baud confirmed functioning.

So, one day the snuffler software can be replaced (or at least partially) with Android code if desired. Or could be used to e-mail or upload the data in real-time so someone can remotely check it while the survey is still being performed.

The test setup was as shown here:

115200 baud confirmed functioning.

So, one day the snuffler software can be replaced (or at least partially) with Android code if desired. Or could be used to e-mail or upload the data in real-time so someone can remotely check it while the survey is still being performed.

The test setup was as shown here:

Hi Guys,

I've been reading your plans on Element14 for an Archaeology Resistivity Meter after seeing it on Facebook.

Its very exciting. Have you considered using a Raspberry Pi unit? I don't know much about the electronics but you can get a GPS unit to fit it. The Pi will also run QGIS software which is 'A Free and Open Source Geographic Information System' which runs on Pi's and Windows Laptops. I have some experience of adding features to QGIS using Python. You could use QGIS to show your results, your position and the grid in the same way as 'High End' Gradiometer Carts do. (Carts can cost £40K or more.) You could use QGIS to display results on a grid, or map, in real time.

I hope this helps.

Paul D

The Pi has been suggested on a couple of occasions.

By the time you have built a Pi and a display and the required power supply and other goodies need to run a serious graphics type application you end up with something rather big and quite power hungry.

For some applications that's no problem at all but for others it is.

The current plan is that the basic box will have interfaces that will allow it talk to anything - USB, logic level UART, BlueTooth and possibly WiFi.

It will have its own controls and display options (simplest display being 4 lines x 20 characters, maximum 800 x 480 pixels with capacitive touch.)

With a simple display I would expect it to use less than 1W (averaged over a day's work).

There would be no problem making it talk to a Pi or anything else.

We haven't got to talking about control software architecture yet but the plan in my head would have the core resistance measuring engine controlled by a simple protocol that can be exposed either to an internal GUI or to the external interfaces.

A Pi would be able to use an external interface to fully control the instrument.

The idea is to rule nothing out but still make the basic system simple enough to get going reasonably cheaply and quickly.

I'm assuming that it will all be open source so access to protocols won't be an issue.

I've come at this from the electronics side of things so QGIS is  new to me - I'll take a look.

It would be great to have Pi based data processing software route and any help in that respect would be massively welcome.

MK

Hi Paul,

I like using Python too. It will be awesome to see what code could be used or developed for this project.

This is a rough sketch of some (not all) of the connectivity methods and how they might be used:

GPS expansion will be feasible by plugging the GPS module directly into the Pi, but it might become unnecessary, since there's the possibility of either sending GPS position from a phone (via Bluetooth or BLE) or sending the measurements into the phone. The only limitation is what's coded, since the underlying hardware covers lots of popular interfaces (USB, BLE, etc). There's memory on-board too, so in theory the data is reliably stored and the SD Card is for transferring stuff (edit: there's no SD card option, it's unnecessary since there are better wired and wireless options) . Also, if it's more comfortable for someone to analyze remotely from a larger computer, then any captured data can be sent in real-time from the phone - assuming it is coded!

I wasn't thinking of supporting an SD card interface, unless others feel this to be essential:

1) The low power Cortex M4 processor doesn't support it

2) If we change processor to one that does we'll need probably to go to 144 pin chip

3) The code to support SD card is complicated

4) You have to support a full on disc filing system

5) Could use a support chip as Gene suggested but it will use a lot of pins.

6) All the above if we support USB OTG except the low power processor is OK with it.

Probably not clear on my original block diagram that the options in the dotted processor outline might need to use a different processor.

On the other hand if we go for a more powerful processor we do get the benefit of faster processing all the time. The H7s support

dual voltage SD card at up to 104MHz - there is some free software from ST which seems to support FAT.

An H7 will burn more power speed for speed and about 300mW more at full (400MHz) speed.

See ST app note AN5200

I'm agnostic on this one - I've got quite used to STM32H7s now

MK

Oh, I see.. sorry I'd blindly copied that and not noticed the dashed line. I'll remove it, I cannot see it ever being needed, since there's more flexible wireless and wired options to transfer data.

A lot of the hand drawn lines look a bit dashed.

I was just warming to the idea of an STM32H7 ...... but I think lower power is better really.

MK

Shabaz,

I have changed your outline. In the UK two of the probes are 'fixed' and can be in the same position for the day (or longer)

The QGIS team have vast experience to call on in displaying this type of data  Many groups in the UK use QGIS and some run training days in Python.

You can have a table and power at P2C2 and use existing equipment (including a tea and coffee facilities - YES most groups have a tent that could be pitched close to the 'fixed' probes if needed) My editing of your original is very bad, sorry.

When I said "What about a Pi" I meant "QGIS would be good. it runs on Laptops and a Pi"

Paul D

Hi Paul,

I've updated it to incorporate your comments regarding the probes and software. One underlying difference (won't make a difference to the user apart from a major improvement hopefully!) is that the older existing designs use analog circuitry for most of the signal generation and processing. In this design however, more of that is done digitally, and so needs precise timing and regular fast calculations, which would be an overhead for the Pi, so all that underlying processing needs a microcontroller (and an FPGA), hence the separation between that and the Pi. The Pi has full control via USB Serial, which is a key difference to Snuffler. The Pi can instruct the desired frequency, start/stop measurements and acquire the data in real-time, i.e. the Pi is in full control using Python, C or any desired language. Also, the interface can still work with existing software in real-time (I've run Snuffler to examine what data is expects), so I've listed those in the diagram now.

For those who want to use a laptop, they will find it equally seamless with Windows and existing software, it's just more convenient to use USB rather than have the older RS232 DB9 connector. Just like with the Pi, they can have full control using any programming language if desired.

For those who want to use Android, initially there could be just a simple skeleton app which doesn't do much but provides some basic visuals, or just collects up the data to e-mail it.

Shabaz,

What software are you using to create your drawings? They are great.

Paul D

Hi Paul,

Thanks! I'm using Powerpoint. I've used it a lot and so already had things like the Pi and display ready to use.

For items like the phone graphic, I usually google search for (say) 'phone transparent png' and then it finds stuff with the transparent background.