MINIDIVER (minipuls+minediver)

[Max]

Max ,can replace 4538 with 4528 or not ? and how must calculate timing ?

Hi,
why not ?
They are too dual monostables... etc like 4538.

Formula is like this:

T(uS) = R*C

so say 100Kohm and 1000pF gives you about 100uS

Not exact of course, look at the picture below.

Kind regards,
Max

[sisco]

Can draw wave form of out 1 and out 2 if:

R1=15K
C1=330PF

R2=10K
C2=3.3NF

R3=10K
C3=47NF

R4=10K
C4=3.3NF

I try understand by PROTEUS software but can't.

(I see in one site, for calculate time must = R * C * ln2)

[Max]

Can draw wave form of out 1 and out 2 if:

R1=15K
C1=330PF

R2=10K
C2=3.3NF

R3=10K
C3=47NF

R4=10K
C4=3.3NF

I try understand by PROTEUS software but can't.

(I see in one site, for calculate time must = R * C * ln2)

Hi,
R*C*ln2 for 74121 or the like not for 4538/4528

they are not the same thing

check again...

Kind regards,
Max

[sisco]

this circuit use in Tiny metal detector and I think this is Determinative delay

time . (all of these metal detectors like tiny , minipuls , minediver not sensitive

to gold and by change this timing maybe can do this)

[Max]

this circuit use in Tiny metal detector and I think this is Determinative delay

time . (all of these metal detectors like tiny , minipuls , minediver not sensitive

to gold and by change this timing maybe can do this)

Hi,
if you change first delay to e.g. 15uS you'll detect gold easy, also small targets. Problem is that at this low delay you detect also what you know.
Big problem for me.

Best regards,
Max

[sisco]

this is minipuls wave form , wave 1 and 2 for pulse width from high power to

low power turn on MOS-FET transistor and 15us before end , out1 (4528 is

on (wave 3).

how must calculate this 15us

T1 = R1 x C1 =>15K x 330p = 3us
T2 = R2 x C2 =>10K x 3.3n = 23us

TOTAL time must T1 + T2 = 3 + 23 = 25 us (may be this time for

resistors or capacitors tolerance)

[sisco]

this is out coil waveform (two head of diode D4 before peramp in minipuls-3)

where of this wave (19us) must use delay for sampling.

(for example gold 15us.this 15us must start of where?)

if I understand good this maybe I can write simple program for ATmega8 for

metal Discernment.

[Max]

this is minipuls wave form , wave 1 and 2 for pulse width from high power to

low power turn on MOS-FET transistor and 15us before end , out1 (4528 is

on (wave 3).

how must calculate this 15us

T1 = R1 x C1 =>15K x 330p = 3us
T2 = R2 x C2 =>10K x 3.3n = 23us

TOTAL time must T1 + T2 = 3 + 23 = 25 us (may be this time for

resistors or capacitors tolerance)

Hi Sisco,
first two waves are tx pulses for big and low power , as I understand.
Ok.
You need to trigger the monostable for first delay using the mosfet control voltage pulse (the tx wave you show as first 2 signals).
You have to trigger 4528 using the falling edge of tx signal at gate of mosfet, not from coil as you made drawing.
Wire clear high, A input low (gnd) and use B input as input for that signal.
The monostable will give you an about 15uS with R=15K and C=1nF
You could use a trimpot of 50K in series with a 10K resistor to gain absolutely right 15uS measured from scope, so avoiding tolerances jitter.

Best regards,
Max

[sisco]

please read my last post ,edit this.

[Max]

this is out coil waveform (two head of diode D4 before peramp in minipuls-3)

where of this wave (19us) must use delay for sampling.

(for example gold 15us.this 15us must start of where?)

if I understand good this maybe I can write simple program for ATmega8 for

metal Discernment.

Hi,
in PI designs the first delay start from the falling edge of gate mosfet triggering signal (convention is at 10% of its magnitude).

If you have a fast monocoil you can sample even at 10uS or less (I reached 8uS)... you need the signal to stabilize fast after switchoff... so low capacitance mosfet , coil , cable etc.

But 15uS is not so difficault to obtain.

Download the bbsailor article and read it:

http://geotech.thunting.com/pages/me...s/FastCoil.pdf

Very good document.

Best regards,
Max

[sisco]

see I true understand:

first delay start bout 8us after turn off MOS-FET (in tiny if I good understand,

start first delay after (R1 x C1) 3us).

delay pule width how ? (I think in tiny 15us)

[sisco]

after reading 4538 and 4053 datasheet I draw this waves .

true or faul ?

I think we need only these two 23us .

first delay start at 3us after pulse , with 23us witdth and second delay start

after 32us at end of firat delay , this is true?

what best timing (for gold detection)? (first delay pulse time , first delay pulse width , second

delay pulse time and second delay pulse width )

for this four timing what is best ?

(with my good english can understand what i say ?)

[sisco]

why is this big difference between delta pulse and tiny about timing:

............tiny.................................. ................Delta puls

start first delay after 3us.......................start first delay after 225us-975us (pot 50K , SIGNAL pot)

first delay pulse width 23us....................first delay pulse width 40us

secand delay start after 32us.................second delay start after 1ms-8.5ms (pot 500K , SENS pot)

second delay pulse width 23us................second delay pulse width 40us

Max these is true!!!!!!!!!!!!!!!!!!!!

what is deltapulse frequncy?

(If I understand gooooood this objects I think can make digital metaldetector with metal Discernment . what you think ?)

[Geo]

why is this big difference between delta pulse and tiny about timing:

............tiny.................................. ................Delta puls

start first delay after 3us.......................start first delay after 30us-110us (pot 50K , SIGNAL pot)

first delay pulse width 23us....................first delay pulse width 45us

secand delay start after 32us.................second delay start after 120us-600us (pot 500K , SENS pot)

second delay pulse width 23us................second delay pulse width 45us


what is deltapulse frequncy? 100Hz

(If I understand gooooood this objects I think can make digital metaldetector with metal Discernment . what you think ?)

Hi Sisco . I marked with red the true clocking of DP.
The differences are so big because DP is a low frequency Hight power for large objects (with large coil). Small objects (especially gold) needs small delay .... and so on........
Regards

[Max]

why is this big difference between delta pulse and tiny about timing:

............tiny.................................. ................Delta puls

start first delay after 3us.......................start first delay after 225us-975us (pot 50K , SIGNAL pot)

first delay pulse width 23us....................first delay pulse width 40us

secand delay start after 32us.................second delay start after 1ms-8.5ms (pot 500K , SENS pot)

second delay pulse width 23us................second delay pulse width 40us

Max these is true!!!!!!!!!!!!!!!!!!!!

what is deltapulse frequncy?

(If I understand gooooood this objects I think can make digital metaldetector with metal Discernment . what you think ?)

Hi Sisco,
Geo already answered to your question.

I dubt that tiny could use 3uS as first delay... never seen anything under 7uS that works. Maybe you've made some error measuring it or used a wrong reference.

Original DP design used a frequency around 100Hz... for XR-71 , that's about the same thing... little differences, frequency is around 400Hz (if I remember well).
That's why these machines were intended for depth search of big objects at meters underground.

DP have a lower frequency cause mounted a large coil as default, XR-71 had more frequency cause was a mix-use intended machine... used also in civil engineering to find underground pipes etc.

When you design a PI for detecting small objects you need a short first delay (usually 15uS or less), then the first sample occours (often width from 15 to 45uS), then the inter-delay that is something great (>100uS, often 200 or more), then the second sample (same width as first sample so 15-45uS).

In a modified e.g. delta pulse you could increase frequency to 400+ Hz (even 1KHz or more)... then reduce the TX-on time to e.g. 50-100uS, giving you a duty factor of 5-10%... that's good without wasting too current for nothing.

Good timings could be then something like these:
f = 1KHz
TX-on pulse width 100uS
duty-factor 10%
1st delay 15uS
1sample width 45uS
interdelay 200uS
2sample width 45uS

That way you'll get small gold items with a good fast monocoil say 8'' or 10'' round.

Everything, gold rings, silver etc will sound good and of course iron !
Also some thick foil.

But then you'll get same problems that you know with iron oxides or ceramics... then is a big problem where there are these things.

Best regards,
Max