Author Topic: Current control capacitive load  (Read 1475 times)

0 Members and 1 Guest are viewing this topic.

Offline MitaliusTopic starter

  • Contributor
  • Posts: 12
  • Country: ru
Current control capacitive load
« on: December 11, 2017, 10:16:17 am »
Hello.
I'm making a medical device that must be attached to the patient.
CHANNEL_ISET changes its state to 100kHz.
If I put the resistor that signal is good. But when I connect the way to the body of signjal becomes sawtooth wave.
What can I fix in this scheme? Any ideas you have on this?

 

Offline DickCappels

  • Newbie
  • Posts: 1
  • Country: th
Re: Current control capacitive load
« Reply #1 on: December 11, 2017, 10:52:00 am »
Differentiate your signal before applying it to the current source circuit.
 

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: Current control capacitive load
« Reply #2 on: December 11, 2017, 10:55:41 am »
The leads will have inductance, the human body will have capacitance, and there will be some resistance and a few other parasitic capacitances, equally some parts of the skin act line non linear resistances,

This circuit at a guess is intended to AC measure the resistance of a person,  instead you may wish to use something like a howland current source, with a dash of compensation so that any inductance in the wires, and capacitance in the person cannot cause ringing or signal degradation.

something like this: http://tinyurl.com/yc2vnt9t

or full link for later readers.

Code: [Select]
http://www.falstad.com/circuit/circuitjs.html?cct=$+1+5e-9+45.7144713268909+58+5+50%0Aa+256+160+368+160+0+5+0+1000000+2.499914676267427+0.7934400248063249%0Aw+368+160+368+96+0%0Aw+368+160+368+224+0%0Ar+256+224+368+224+0+10000%0Ar+256+96+368+96+0+10000%0Ar+256+96+144+96+0+10000%0Ar+144+224+256+224+0+10000%0Aw+256+176+256+224+0%0Aw+256+96+256+144+0%0Ar+256+320+256+400+0+47000%0Aw+256+400+304+400+0%0Ag+256+400+256+432+0%0Aw+256+320+304+320+0%0Ax+187+67+217+70+4+24+R1%0Ax+300+67+330+70+4+24+R2%0Ax+185+259+215+262+4+24+R3%0Ax+298+259+328+262+4+24+R4%0AR+144+96+80+96+1+2+100000+2.5+2.5+0+0.5%0Ac+304+320+304+400+0+1.0000000000000001e-7+0.7934299896949569%0AO+368+160+480+160+0%0Al+256+320+256+272+0+0.00009999999999999999+0.000059137027491623215%0Ar+256+272+256+224+0+1%0Ac+256+176+144+176+0+1.0000000000000001e-7+-0.2762715787593846%0Ar+144+224+144+176+0+1000%0Ax+317+333+518+336+4+24+Skin%5CsAprroximation%0Ax+43+145+237+148+4+24+AC%5CsCompensation%0Aa+48+224+144+224+1+15+-15+1000000+1.0629814961692903+1.062992125984252%0Aw+144+224+144+272+0%0Aw+144+272+48+272+0%0Aw+48+272+48+240+0%0Ar+-32+208+-112+208+0+10000%0Ag+-32+272+-32+288+0%0Aw+-32+208+48+208+0%0AR+-112+208+-128+208+0+0+40+5+0+0+0.5%0Ar+-32+208+-32+272+0+2700%0Ax+0+296+46+299+4+24+Bias%0Ao+19+64+0+12290+1.8882847784322407+0.0001+0+1%0A
 
The following users thanked this post: Mitalius

Offline Rerouter

  • Super Contributor
  • ***
  • Posts: 4694
  • Country: au
  • Question Everything... Except This Statement
Re: Current control capacitive load
« Reply #3 on: December 11, 2017, 10:59:27 am »
To better clarify, the point between R3 and R4 would be your output for your electrode, where it says out would be the measurement output.

The AC compensation values i pulled off the top of my head based on the range of inductance and capacitance I know myself to measure on average.
 
The following users thanked this post: Mitalius

Offline MitaliusTopic starter

  • Contributor
  • Posts: 12
  • Country: ru
Re: Current control capacitive load
« Reply #4 on: December 11, 2017, 11:40:55 am »
Thank
I'll have to try this in the next version of the board.

May I know the formula for Iout?
« Last Edit: December 11, 2017, 01:31:46 pm by Mitalius »
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14069
  • Country: de
Re: Current control capacitive load
« Reply #5 on: December 11, 2017, 05:21:31 pm »
The given circuit is a kind of current source. So it tries to give an output current of U_in / 200 Ohms. With a capacitive load and a square wave drive level with would ideally be a triangle wave for the voltage. So the circuit is not that far off.  100 kHz is rather fast for such a circuit - so not all parts be behave like ideal resistors and OPs. Especially the OPs will be less than perfect. The extra phase shift from the second OP might even lead to instability. Not so sure square wave is a good idea at 100 kHz - it might become difficult to meat FCC standards.
 

Offline MitaliusTopic starter

  • Contributor
  • Posts: 12
  • Country: ru
Re: Current control capacitive load
« Reply #6 on: December 12, 2017, 09:46:26 am »
Maybe I can use coax?

Anyway, I'm going to have to here the right form of signal first.
 

Offline danadak

  • Super Contributor
  • ***
  • Posts: 1875
  • Country: us
  • Reactor Operator SSN-583, Retired EE
Re: Current control capacitive load
« Reply #7 on: December 12, 2017, 12:50:51 pm »
You look like the signal is slewing at its rated value, maybe 10 V / uS but
hard to tell without expanded scope trace. This on rising edge, but falling
edge 1V / 5 uS. Begs the question what does the input signal look like ?

Do a scope capture of 2 cycles input and output, and post on forum.


Regards, Dana.
« Last Edit: December 12, 2017, 12:56:20 pm by danadak »
Love Cypress PSOC, ATTiny, Bit Slice, OpAmps, Oscilloscopes, and Analog Gurus like Pease, Miller, Widlar, Dobkin, obsessed with being an engineer
 

Offline Lee Leduc

  • Regular Contributor
  • *
  • Posts: 84
  • Country: us
Re: Current control capacitive load
« Reply #8 on: December 12, 2017, 01:02:43 pm »
Google "biomedical howland current source" and you will find lots of references for this circuit. For example "Biocompatible, High Precision, Wideband, Improved Howland Current Source With Lead-Lag Compensation"
http://ieeexplore.ieee.org/document/6221965/
 
The following users thanked this post: Mitalius

Offline MitaliusTopic starter

  • Contributor
  • Posts: 12
  • Country: ru
Re: Current control capacitive load
« Reply #9 on: December 20, 2017, 08:35:20 am »
I have emissions how can I get rid of them?
And I still don't know the formula Iout my scheme
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf