Stuart/Paul II PI revisited.

[ericwd9]

Despite all the problems that this project has had, the basic design is really very good. Using the Magenta PIC code it was an acceptable detector. Mark Paul suggested revisions that would have made this design a real killer. Unfortunately the .asm file commented in dansk would assemble a hex file that would operate (with some difficulty) but would not save to eeprom. With a lot of help from Maxime Thibault to sort out the code and improve it and a few modifications to the scematic this project is a go. It will allow real time changes to Pulse width, Pulse repetition rate, Sample width and Sample delay. Then it will store these parameters in 4 stores for recall. (Four setups for four coils?) I intend to incorporate an LCD oscilloscope ($39.00 ebay) This will facilitate in the field setup to optomise operation. Mods are needed to prevent the mosfet from conducting before the PIC is operating as this causes excessive current. Those who have failed to have the PIC code as previous work at all, need to know that without mods the coil should be disconnected at startup. Further the PIC at first startup has not stored the parameters in eeprom. It will give an excessive pulse width. You must start first time with micro-switch 4 closed to store these values. Most of the previously provided firmware will operate this way. Those (FIRMLY) interested can contact me for upgrades as they occur. This was a good design bedeviled by corporate greed and poor programming skills from builders.
Eric.

[WM6]

Hi Eric

Thanks for mods info.

What mods are needed to prevent the mosfet from conducting before the PIC is operating as this causes excessive current?

[ericwd9]

I don't have the schematic with me right now. But change the 10K res. from the PIC to the drive transistor to 1K and add a 10K from the PIC to +5V supply. This holds the mosfet in the off state until the PIC starts up. Another mod: Change the 1000 uF to 2500 uF I will write up the mods properly when I have finished and have time. Also the PIC asm and Hex files are still being tidied up. Some changes to the coil damping might be needed depending on the coil and the use it is put to. Also my new codes are writen for PIC16F84A. Also 2 further pins of the PIC are used to designate the four memory stores and read write for them. More will be forthcoming when results and conclusions are complete. Working on Litz/Basketware fast 8in coil. Off to the WA goldfields in 2 weeks for 3 months so there will be a recess.
Eric

[satdaveuk]

Very interesting
Thanks Eric

[ericwd9]

Hex File for Stewart P.I. II

With thanks to Mark Pauls. for the original assembly code (not working correctly but most of the work done)
With thanks to Maxime Thibault. for many improvements and changes.
When first operated start with micro switch RA4 closed.
Default timing: Pulse duration 20uS. Receive delay 10uS. Received time 5uS. Delay until next pulse 40uS.
Change of timing for each key-press 1uS see: http://www.geotech1.com/cgi-bin/page...art1/index.dat for other details. Note that there are errors in the circuit diagram, That modification need be made and that there is one change to the board needed. Here for those needing it is the assembled Hex code. The source code needs more work and is notated in three languages at present. It is written for 16F84A and contains fuse operand.

Hex File.

:100000005A280B3093000E309200920B0528930B68
:10001000032808000120910B0A2808002F088900F6
:100020002E08880083168B13081555308900AA30D6
:10003000890088148B1703118312643091000A2001
:10004000080083122F0889008316081483120808F9
:10005000AE000800051A35280518970A8518970379
:100060000519960A8519960339280518950A8518E1
:10007000950B1508AE000130AF000E201608AE003B
:100080000230AF000E201708AE000330AF000E2084
:1000900003308C00FF3091000A208C0B4A280800A6
:1000A000143095008A00960005309700283098009B
:1000B00039208828FF30860083160030860083129E
:1000C0000030850083161F3085008312051A5028E2
:1000D0000130AF0021202E0895000230AF00212012
:1000E0002E0896000330AF0021202E0897002830FC
:1000F00098008828150890000610900B7D2806149B
:100100000800170890008611900B842886150800B7
:100110007A2016089000900B8B28812018089100F7
:100120000A2005182A2085182A2005192A20851951
:040130002A208828D1
:02400E00F13F80
:00000001FF

Note that ver2 will have 4 stores for storing parameters for more coils or different search needs. This is ver1
I am off to the W.A. goldfields where field test will take place. I will be back in Nov. Enjoy!

Eric

[WM6]

Thanks for all your efforts and kindness Eric and
more gold than you can bring with from W.A. goldfields.

[ericwd9]

List of harware modifications Stuart PI II

There are a number of errors in the schematic diagram that are not carried to the board layout. A number of connections that should go to the +16V are shown going to +9V. The board is OK here.
Two connections on the Board need changing. VR1 should go to +16V not +9V. R3 should go to pin 6 and not pin 7 of IC2A. Change R22 from 10K to 1K. Add a 10K resistance from pin 6 of IC5 to +5V. (this is to keep TR4 non conducting until IC5 starts up.) Change C16 from 1,000uF to 2,500uF. Add a 0.33 or 0.47 metalized Cap across C16. Add 50K cermet pot, connected to +16V, 0V and wiper to R5 removing the connection of R5 from the + supply. (this is coarse balance adjustment) Adjust this to center VR1. Take care with TR1, orientation on board layout is wrong, schematic is correct. Note that BC213 has differing pin-outs to BC213L. R26 and R27 will need tailoring to your coil. I used 3.3K R27 and 360R for R26 (2 x 180R) Note that this places the diode D3 conducting a big slice of the coil current. At 20uS this is not any problem but should you have a much longer Pulse duration D3 should be replaced with a higher current, high speed recovery diode. I have not done this (yet) A high speed diode could be put in circuit from the drain of TR4 to the coil, Cathode to drain. This will negate much of the mosfet capacitance and speed up the coil some, requiring a larger value of R26 and R27. I have not done this (yet) This will have your MD at the same stage in development as mine. You could add 2 more micro-switches with resistors as for RA0-RA4 on RB4 and RB5 of IC5. These will be used for memory + and recall in software ver2 (in development) I do not have the time right now to write all this up as I should and correct the schematic and the board layout but this description carried out will suffice for now. Note that the first time you power up you must have the coil disconnected and switch RA4 closed. (This is so that the pulse values are written to the EEPROM of the PIC.) If this is not done the values will be in mS instead of uS. The long pulse duration could fuse D3 if the coil is connected. Once the first start-up is complete it will always start-up in uS. Closing RA4 on start-up will load the defaults if you get lost with fine tuning the settings.
Eric

[bernte_one]

i have another one
for all free
pi84.txt

[ericwd9]

i have another one
for all free
[ATTACH]36102[/ATTACH]

What are the defaults on this code?
Eric

[bernte_one]

i share out my archive
but forum space is not big enough, i have all files for latest stuart including hex, just write me

[BLACKSEA]

plaese my send yourself mail, stuart2

[WM6]

plaese my send yourself mail, stuart2

[BLACKSEA]

HELLO WM6

[iw5dxh]

hello bernte_one, can I have too the stuart2 documentation, please? thank you!