using LF 198 , help

[okantex]

Hi ,


is anybody knows how to use LF198 , you know it is used in PD 3200 .

I want to learn how does it works? , is anybody who had used it before.

I could not understand how to use it form it's manual.


and for PD3200 , why there is no drain to ground between LF198 output and Atmega's ADC input.
there is one drain between LF198 and LM356 ,but why there is no one after LF198?

regards
okantex

[WM6]

look at tipical application

http://www.datasheetcatalog.org/data...ogy/lt0198.pdf

[Max]

Hi ,


is anybody knows how to use LF198 , you know it is used in PD 3200 .

I want to learn how does it works? , is anybody who had used it before.

I could not understand how to use it form it's manual.


and for PD3200 , why there is no drain to ground between LF198 output and Atmega's ADC input.
there is one drain between LF198 and LM356 ,but why there is no one after LF198?

regards
okantex

what do you mean for drain ?

look here...
http://mysite.du.edu/~etuttle/electron/elect53.htm

It's sample and hold... look datasheet carefully, it's easy to use.
You must consider it has digital ground and mcu must provide digital s/h pulse according to typical timing of device. You have to choose carefully external component otherwise you'll get wrong readings.

An important issue is that device has hi-impedance input so care must be taken to not damage it handling or soldering etc.

Kind regards,
Max

[okantex]

first of all thanks for interests ,

Max ,

look at the link below
http://www.geotech1.com/forums/showt...=15082&page=11
see schematic at post 259 ,

you will see connection line marked as PC0 which goes to ADC of chip directly.

you will see D6 and C10 parallel to eachother , I call it drain , I am sure I am wrong about it.
I thought that it is for protecting chip's ADC from unwanted signals.is not it a kind of drain for ADC? if you want to get sample you hold on ADC pin ,otherwise signal flows from D6 and C10 ,am I wrong?

I wonder ,when usin lf198 why do not we do the same after output of lf198 ? I mean between 198 and ADC pin.

thanks for the link I read it , does the output decreases only by leakage of hold cap? how long does ,it takes ? do I have to wait it? I think no ,If I change lf198's pin 8 agin to "0" will the output start to show ( follow ) input value immadiately ?
in the link it says 28 min. to become zero volt.

regards

[okantex]

btw ,

does LF198 change polarity of signal.?

can I add this lf198 directly to above schematic ,
I want to add 2 lf198 to other ADc pins which are not being used without canceling direct connection at above schematic

is it possible?

[Tinkerer]

btw ,

does LF198 change polarity of signal.?

can I add this lf198 directly to above schematic ,
I want to add 2 lf198 to other ADc pins which are not being used without canceling direct connection at above schematic

is it possible?

okantex,

D6 and C10 are there to shift the level of the output of the opamp. The opamp has a +5/-5V supply and the MCU has 0/+5V supply.

The output of the LF198 will be the DC sample taken while on Hold (0V) and will take a new sample when on Sample (5V)
Look at the TINKERERS circuit at:
http://www.geotech1.com/forums/attac...2&d=1245941313

where you can see LF 398 in use. It is the same as LF198. There you also see an PL filter at the output and an averaging RC filter on the Hold pin.

all the best

Tinkerer

[okantex]

Hi Thinkerer ,
how are you , thanks for your help

look at below link you will see pd3200 schematics there ,
http://www.teknolojiekibi.com/yeni/i...p?topic=2053.0

A) there in pd3200 C11 and R43 are doing same operation with D6 and C10 of 1843 's PI111 . Am I right.

B ) I want to ask you a question about your circuit , what is the use of C13 ,C7 ,C17 or C15 at your circuits after Lf398s ??
are not they a kind of drain to divert unwanted signals.?

C ) at Pd3200 sample hold cap is 4700 ,is not it 4700uF ?
why it is 4700uF , must not it be something at level of nanofarads maybe 4,7nF for 10us sampling time???

D ) why there is nothing llike C7 of your circuit after PD3200's LF198 , does not it need to give a link to AGND to let it work??
as I know there is not inner Cap to do the same operation at atmega chip's ADC channel entrance.??

regards
Okantex

[okantex]

and what is the uses of your circuit ? what does it detects ,I think it is not an active detector circuit ,looks like a passive receiver.

regards
Okantex

[Tinkerer]

and what is the uses of your circuit ? what does it detects ,I think it is not an active detector circuit ,looks like a passive receiver.

regards
Okantex

Hi okantex,

the TINKERERS signal processing circuit is the RX part of a discriminating PI detector. The TX part is on the power supply drawing. It is not complete there either. One part is with the coils.
Yes, I need to put it all together....

On the PD3200:
R46,1k and C14,4700 (you can try 470n or even 4.7u, it depends on how many uS the sampling pulse is and how many PPS) is an RC lowpass filter, it averages the output of the S&H LF398.
I do the same on the pin 6. Hold.
Pin 5 is the output, it is another RC lowpass filter.

Tinkerer

[Tinkerer]

Hi Thinkerer ,
how are you , thanks for your help

look at below link you will see pd3200 schematics there ,
http://www.teknolojiekibi.com/yeni/i...p?topic=2053.0

A) there in pd3200 C11 and R43 are doing same operation with D6 and C10 of 1843 's PI111 . Am I right.

B ) I want to ask you a question about your circuit , what is the use of C13 ,C7 ,C17 or C15 at your circuits after Lf398s ??
are not they a kind of drain to divert unwanted signals.?

C ) at Pd3200 sample hold cap is 4700 ,is not it 4700uF ?
why it is 4700uF , must not it be something at level of nanofarads maybe 4,7nF for 10us sampling time???

D ) why there is nothing llike C7 of your circuit after PD3200's LF198 , does not it need to give a link to AGND to let it work??
as I know there is not inner Cap to do the same operation at atmega chip's ADC channel entrance.??

regards
Okantex

Answer A:
On the PD3200, the MCU and the LF356 opamp use the same single power supply, so there is no level shifting needed.
C11 and R43 are feedback of the amplifier.

Tinkerer

[okantex]

hi Thinker ,
I think there is misundertood of mine..
can you explain
"R46,1k and C14,4700 (you can try 470n or even 4.7u, it depends on how many uS the sampling pulse is and how many PPS) is an RC lowpass filter, it averages the output of the S&H LF398.
" a little bit more.

why pd3200 has 4700uF.


from pdf file of 198 ,I checked acquisition time graph, it is between 0,001....,01 uF.

I want to mean ,I know sampling time is chosen by MCU but according to this time's duration ,we chose hold cap value which averages the output value.
this cap may effect value of output in relation with sampling time but
I am realy confused now ,
is not 3200's 4700uF to big value.
normal MCU 's ADC sampling times are around 18,47 us minimum , we use lf198 because of it has lower sampling times , even for 20us sampling time according to graph at LF198's pdf we have to use 0,01uF hold cap.
how can PD3200 use it?
you also used 100N at yours


regards

[Tinkerer]

hi Thinker ,
I think there is misundertood of mine..
can you explain
"R46,1k and C14,4700 (you can try 470n or even 4.7u, it depends on how many uS the sampling pulse is and how many PPS) is an RC lowpass filter, it averages the output of the S&H LF398.
" a little bit more.

why pd3200 has 4700uF.


from pdf file of 198 ,I checked acquisition time graph, it is between 0,001....,01 uF.

I want to mean ,I know sampling time is chosen by MCU but according to this time's duration ,we chose hold cap value which averages the output value.
this cap may effect value of output in relation with sampling time but
I am realy confused now ,
is not 3200's 4700uF to big value.
normal MCU 's ADC sampling times are around 18,47 us minimum , we use lf198 because of it has lower sampling times , even for 20us sampling time according to graph at LF198's pdf we have to use 0,01uF hold cap.
how can PD3200 use it?
you also used 100N at yours


regards

Okantex,

I think 4700uF is a mistake it could be 4700nF or even 4700pf.

On the TINKERERES I use sampling times of 3.5uS to 15uS for different samples, but I have already used 1uS samples and it worked fine.
The sampling times indicated on the Datasheet, refer to sample settling time at 10V, when taking a sample starting at 0V. However, with the PI, the difference between the voltages of the samples from one pulse to the other is very small, so the capacitor charge time is short. Only small steps.
On the TINKERERS circuit that you are looking at, I used 600PPS (Pulse Per Second) but I think the PD 3200 uses only 100PPS, so to average the hold voltage you need a larger capacitor.

Tinkerer

[okantex]

dear friend ,
since we are talking about MCU controlled pulsing , there is no meaning of considering frequcies ,

important is the space that is suitable to charge to get average output signal from unit.

since it is mcu controlled unit , one cycle and hundreds cycles will need same stndart hold cap.
if this were a kind of ADC unit and yuo were taking charges at same durations of differents pulses and than getting average value ,maybe you could be right,
but 4700uF is too much to charge ,

did you get 1us sampling time 's sample with 100nF hold cap ?I would expect it with 0,1nF hold cap .

regards