PI CHIP

[mickstv]

Mick check the button inputs for normal operation these should all be pulled up to +5 volts. Pressing a button should pull low. However all should be high for normal operation

Check the supply and grounds to the analogue pins on the chip ... the logic could be working OK but there could be a power fault on the analogue section ... check ref in pins also for 0 and +5 volts.

If you can dump a screen from the serial uart tool for the Pickit .... 9600 baud.

paul.

Hi Paul, I'll check all the connections again the pull up resistors and post back tomorrow. I Don't think the uart tool works with PicKit 3 but I'll check that also.



Mick

[moodz]

Hi Paul, Just need to do some checking I built up a rough detector similar to the schematic you posted using a NE5534, 4066 and a comparator IC anyway all the timing seems to work and I can see voltage change on the output of the comparator but no change on the output of the pic and no signal change on display. Just won't to check that pin 5 is the ADC input as per the schematic you posted.
Mick

Mick check the button inputs for normal operation these should all be pulled up to +5 volts. Pressing a button should pull low. However all should be high for normal operation

Check the supply and grounds to the analogue pins on the chip ... the logic could be working OK but there could be a power fault on the analogue section ... check ref in pins also for 0 and +5 volts.

If you can dump a screen from the serial uart tool for the Pickit .... 9600 baud.

paul.

[mickstv]

Hi Paul, Just need to do some checking I built up a rough detector similar to the schematic you posted using a NE5534, 4066 and a comparator IC anyway all the timing seems to work and I can see voltage change on the output of the comparator but no change on the output of the pic and no signal change on display. Just won't to check that pin 5 is the ADC input as per the schematic you posted.




Mick

[mickstv]

Already on it ... the dpsic30F4011 code is already built ... just needs testing. This adds more buttons and 4 sample pulses S1 S2 S3 S4. I will post some results.

Give the Diff front end a try in surfpi config as a starter ... I was going to go onto that next but happy to work with everyone .... mostly if you find bugs or need more features tell me and I will try to fix. Also opinion on whether the chip should use crystal oscillator instead of internal clock would be good too.

moodz.

Hi Moodz, I should have a rough front end done over the weekend and will post up some pictures and results.



Mick

[moodz]

This particular back end won't work with the Goldscan because I only have two sample points and the Goldscan needs four. I could use it with HH or SurfPI though. Or I could also try Moodz Diff front end then into a 4066 and comparator and see how this Diff front end works.

What does Moodz think of the last idea ?



Mick

Already on it ... the dpsic30F4011 code is already built ... just needs testing. This adds more buttons and 4 sample pulses S1 S2 S3 S4. I will post some results.

Give the Diff front end a try in surfpi config as a starter ... I was going to go onto that next but happy to work with everyone .... mostly if you find bugs or need more features tell me and I will try to fix. Also opinion on whether the chip should use crystal oscillator instead of internal clock would be good too.

moodz.

[mickstv]

Hi Mick
great job
I will be interested to see which front end you select to make,
moodz said somewhere that the Goldscan Circuit works
but I dont have that schematic, the unisurf looks ok to,
but untested.

This particular back end won't work with the Goldscan because I only have two sample points and the Goldscan needs four. I could use it with HH or SurfPI though. Or I could also try Moodz Diff front end then into a 4066 and comparator and see how this Diff front end works.

What does Moodz think of the last idea ?



Mick

[6666]

Hi Mick
great job
I will be interested to see which front end you select to make,
moodz said somewhere that the Goldscan Circuit works
but I dont have that schematic, the unisurf looks ok to,
but untested.

[mickstv]

mickstv
did you get this project working ?

Hi 6666, Yes I've got the processor section finished and programmed I used PICkit 3 programmer. Now all I have to do is build the front end then start experimenting.






Mick

[6666]

mickstv
did you get this project working ?

[moodz]

very small bug fix

version in serial output was hard coded string at v 2.0 now fixed and gets correct version info from config file in source code ...

38 microvolt sensitivity from a 10 bit ADC

>
Enter a command : eg A1234 sets Tx freq count to 1234

MooDz UNIPI v230
S: Save all vars to EEPROM
R: Read all vars from EEPROM
CHECKSUM: RAM = -30752 EEPROM = -20
STATUS:
X: CPU Ref Clock : 4 X 7370.000 Khz = 29.480 Mhz
PWM Step is 33.921 Nanosecond
A: TX FREQUENCY : Count = 7370 Freq = 4.000 Khz Period = 250.000 usec
B: TX PRIMARY PULSE : Count = 0 Duty = 0.000 usec
C: TX DESCRIM PULSE : Count = 230 Duty = 3.901 usec
PWM3 TARGET level : Count = 7371 Duty = 125.017 usec
E: SAMPLE PULSE 1 ON : Count = 0 ON = 0.000 usec
F: SAMPLE PULSE 1 OFF : Count = 0 OFF = 0.000 usec width = 0.000 usec
G: SAMPLE PULSE 2 ON : Count = 65535 ON = 17784.260 usec
H: SAMPLE PULSE 2 OFF : Count = 65535 OFF = 17784.260 usec width = 0.000 usec
K: TARGET PWM scale : 15
ADC sample = 4104 [0.156556 Volts]
ADC SAT sample = 4105 [0.156595 Volts] DIFF = 1 [0.000038 Volts]
M: AUDIO freq range count = 65535
SELECT Button: 1 ZERO Button: 1 SAT LOCK Button: 1
>

[moodz]

YAU ... yet another update dspic 30f4012

fixed floating point error on serial port data ... gave wrong pulse timings.
now OK on LCD and serial.
fixed erratic rotary switch LCD menu selection ... very smoooooth now.

dump from UNIPI serial port console session ...

>
Enter a command : eg A1234 sets Tx freq count to 1234

Universal PI chip by Moodz v2.0
S: Save all vars to EEPROM
R: Read all vars from EEPROM
CHECKSUM: RAM = -24078 EEPROM = -15692
STATUS:
X: CPU Ref Clock : 4 X 7370.000 Khz = 29.480 Mhz
PWM Step is 33.921 Nanosecond
A: TX FREQUENCY : Count = 7370 Freq = 4.000 Khz Period = 250.000 usec
B: TX PRIMARY PULSE : Count = 5896 Duty = 100.000 usec
C: TX DESCRIM PULSE : Count = 230 Duty = 3.901 usec
PWM3 TARGET level : Count = 7378 Duty = 125.136 usec
E: SAMPLE PULSE 1 ON : Count = 406 ON = 110.176 usec
F: SAMPLE PULSE 1 OFF : Count = 440 OFF = 119.403 usec width = 9.227 usec
G: SAMPLE PULSE 2 ON : Count = 599 ON = 162.551 usec
H: SAMPLE PULSE 2 OFF : Count = 699 OFF = 189.688 usec width = 27.137 usec
K: TARGET PWM scale : 15
ADC sample = 4103 [0.156518 Volts]
ADC SAT sample = 4111 [0.156823 Volts] DIFF = 8 [0.000305 Volts]
M: AUDIO freq range count = 65535
SELECT Button: 1 ZERO Button: 1 SAT LOCK Button: 1
>
Enter a command : eg A1234 sets Tx freq count to 1234

Universal PI chip by Moodz v2.0
S: Save all vars to EEPROM
R: Read all vars from EEPROM
CHECKSUM: RAM = -24086 EEPROM = -15692
STATUS:
X: CPU Ref Clock : 4 X 7370.000 Khz = 29.480 Mhz
PWM Step is 33.921 Nanosecond
A: TX FREQUENCY : Count = 7370 Freq = 4.000 Khz Period = 250.000 usec
B: TX PRIMARY PULSE : Count = 5896 Duty = 100.000 usec
C: TX DESCRIM PULSE : Count = 230 Duty = 3.901 usec
PWM3 TARGET level : Count = 7370 Duty = 125.000 usec
E: SAMPLE PULSE 1 ON : Count = 406 ON = 110.176 usec
F: SAMPLE PULSE 1 OFF : Count = 440 OFF = 119.403 usec width = 9.227 usec
G: SAMPLE PULSE 2 ON : Count = 599 ON = 162.551 usec
H: SAMPLE PULSE 2 OFF : Count = 699 OFF = 189.688 usec width = 27.137 usec
K: TARGET PWM scale : 15
ADC sample = 4112 [0.156862 Volts]
ADC SAT sample = 4111 [0.156823 Volts] DIFF = -1 [-0.000038 Volts]
M: AUDIO freq range count = 65535
SELECT Button: 1 ZERO Button: 1 SAT LOCK Button: 1
>

[moodz]

New hex file for v23 dspic30f4012

Fixes ...

1. found a bug in the timer interrupt code .. not serious ... fixed now.

Upgrades ...

1. Code runs twice as fast .. more samples = less noise.

2. 16 bit resolution on target ADC ... min 18 bit on SAT.

3. Auto SAT tracking ... much more stable.


moodz

Metal23hex.zip

[moodz]

Great moodz, this is for us old fashioned boys. Probably new hex file too?

Just testing the code then will release hex file.

[WM6]

UNIPI now moved to slightly bigger chip .... dspic30F4011. Forty pin DIP so easy to solder

Great moodz, this is for us old fashioned boys. Probably new hex file too?

[moodz]

UNIPI Universal PI Controller V3

UNIPI now moved to slightly bigger chip .... dspic30F4011. Forty pin DIP so easy to solder

4 sample pulse lines now .... works in Goldscan Circuit ...
Each pulse line can have individually settable ON and OFF times during or after TX pulse in 66 ns step resolution.

Dynamic control of coil damping. Optimise coil damping at the press of a button.

Crystal Clock option.
More buttons.
Real Serial port.

All features in UNIPI v230 still included.