# DC block

Discussion in 'Electronic Design' started by john, Nov 13, 2007.

1. ### johnGuest

Hi,

I am using the following circuit ( Figure: 16 , page 13 )

I connected the INA 133'S pin#3 to the output of the DAC (voltage
out). DAC outputs DC voltage of minus 2.5 volts at the power up.
Inorder to avoid DC -2.5volts , I put a capacitor in between the DAC's
output and the non inverting inout of the INA, but I am still getting
-2.5 volts at the non inverting input of the INA. Can somebody advice
whats really happening? Capacitor suppose to sblock DC.

Thanks
John

2. ### Tim WescottGuest

Oy. Where to start...

Capacitors don't _conduct_ DC, and they only _block_ DC in the sense that
they insure that no DC current flows. If the amplifier pin wants to drift
to -2.5V, then the blocking capacitor will just insure that that happens
all the time.

I suggest you find some place to post your whole circuit, or at least the
part that includes the DAC and amplifier, and post it over on
sci.electronics.basics.

--
Tim Wescott
Control systems and communications consulting
http://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes, http://www.wescottdesign.com/actfes/actfes.html

3. ### Jan PanteltjeGuest

It is wired as voltage follower, input impedance at pin 3 extremely high.
The cap will not charge if no current flows.

4. ### BobGuest

We need to see the whole circuit to see what is going on.
What value is the capcitor connector to pin 3?
What have you done with pin 2 on the INA2133?
What value of R (as shown in figure 16) are you using?
What are the power supply voltages?
Do you have the DAC ground connected to a 0v rail in between
the pin7 +v and pin 4 -V rail?

I don't think you are giving us the full story.

Assuming the problem really is that the DAC output has -2.5V
DC on it then you need to consider the switch on behaviour.
The capacitor connected to pin 3 then it
will take some time to charge up so there is 2.5V across
the capcitor. The capacitor charging current is limited
by the 25K resistor inside the INA2133
For example a 220uF capacitor will take 5.5 seconds to
charge up to 2.5V from a 0V initial charge.

Bob

5. ### johnGuest

Hi,

1. What value is the capcitor connector to pin 3?
470uF

2. What have you done with pin 2 on the INA2133?
Grounded

3. What value of R (as shown in figure 16) are you using?
5kohm

10kohm

5. What are the power supply voltages?
+/- 15volts

6. Do you have the DAC ground connected to a 0v rail in between
the pin7 +v and pin 4 -V rail?
Yes

7. It is wired as voltage follower, input impedance at pin 3 extremely
high.
The cap will not charge if no current flows.

I also connected a 1kohm resistor on the pin 3 after the capacitor and
grounded it but still the INA is getting -2.5 volts.

Regards,

John

6. ### Jan PanteltjeGuest

Shorted cap, or you measured at the wrong side of it?

7. ### johnGuest

Shorted cap, or you measured at the wrong side of it?

Cap. is not shorted and measured it on the right side.

john

8. ### Jan PanteltjeGuest

OK, you gave that diagram, and I looked it up first time.
You say you connected it to ground.
When and while you did, you MUST have read 0v?
Else remove the cap, and measure voltages again?

9. ### johnGuest

I removed the resistor and just used the capacitor, I am measuring
-2.5volts at the input of the capacitor and -10 volts at the output of
the capacitor or at the input of the INA. SO, rigth now the setup is
DAC ------ capacitor----------INA (PIN3)

John

10. ### Jan PanteltjeGuest

I am still lookin gat that diagram with the OPA131,
What is on pin 2?

11. ### johnGuest

Pin2 of INA is grounded. The problem has changed. I want to post the
circuit diagram. how can I do it?

John

12. ### Jan PanteltjeGuest

There are several sites where you can upload a picture, I have my own server,
so I cannot tell you which one

13. ### johnGuest

Hi,

http://profile.imageshack.us/user/ai466/images/detail/#69/circuitdiagrambd1.png

The problem is that at the power up, - 0.2 volts appears across the
load ( 10kohm) which is not desriable for my application. The load is
isolated by a 10uF capacitor. The voltage at the input of the
capacitor is -2.4 volts ( at the power up) . I do not know that why
this capacitor is not blocking the DC voltage completely. I tried
polypropalene 0.5 uF capacitor and the voltage across the load drops
to 0.0001mvolts ( desirable) . But the input voltage at the capacitor
increased from - 2.4 volts to -14 volts.

The capacitor 470uF is not blocking the -2.4 volts either, its output
is -0.5 volts going to INA's pin 3.

John

John

15. ### Jan PanteltjeGuest

OK I got it.
The circuit is a voltage to current converter.
The output current will at first flow in the 10uF capacitor,
but after that capacitor has charged, only in the 1MOhm load.
Even a VERY low current in a 1MOhm load, will create a huge output voltage!!!!!
The output voltage comes back to the input via the resistors in the INA133.

Nothing to do with those capacitors (but watch the polarity).

What you perhaps COUKLD do (or I would do), is short the 10uF, remove the 1MOhm,
short the 470 uF, remove the 385 Ohm, and apply a volatge of say 2.5V DC
from some reference at pin 2 of the IN133.
That way the DC path is correct, and the output will be (UADC - 2.5) / R.
Select the 2.5V on pin 2 so you get the right swing.
For example for a 0-5V ADC, output current would swing from negative to positive
with 2.5V on pin 2.
No capacitors needed.
The 2.5 must come from a low impedance point.

Well, it is late here, but something like that

16. ### johnGuest

Hi,

What should I do about the leakage current? The put the 10uF capacitor
to stop leakage current.

John

17. ### Jan PanteltjeGuest

I am not sure what you mean.

If pin 2 and pin 3 of the INA133 are at same potential,
then the output current (without capacitor) in a *low impedance*
load will be (V3-V2) / R, so zero!

I marked *low impedance* as lets take a numeric example:

Say pin3 is +1.5V (some AD ouput).
and pin 2 is +2.5V (the reference).

Now if the load is a short-circuit (you amp meter),
then the output current will be (1.5 - 2.5) / 5000 = -1 / 5000 = -200uA.
If you made the load impedance 100 kOhm then that would cause a voltage
drop of -.0002 x 100000 = -20V, and your circuit CANNOT supply that.
So you load impedance has a maximum value where it will still work.

I do not know why you want a current source, or what your load is,
but that is one factor.

So if your DA conbverter has for exampe an output from 0 to +5V,
YOU have to decide where the zero point is of Iload,
and you can do that by setting the apropriate voltage on pin2.
With ADC midrange 2.5V and 2.5V on pin 2, there is zero output current.
If you put 0V on pin 2, then 0V ADC out gives zero output, and 5V gives
1mA output.
If you have say 10V supply, then the max Rload is 5kOhm.

If pin 2 is at 5V, the 0V on the ADC wil lresult in -1mA output,
while 5V on the ADC will result in 0mA output.
So the voltage on pin 2 sets your range, the 5kOhm your gain.

Hope I made no mistakes, but do you get the idea?

18. ### JamieGuest

Well, I'll guess. did you put the cap in backwards assuming it
maybe polarized?
did you confirm the INA outputting 0 volts for example with no
input?
You may want to ding around a bit here.

19. ### JamieGuest

You need to use DC coupling with offset biasing to obtain your
0.0 output..
Using caps will just add to the problem.

Also, the settling time on the DAC from the process control end
will place a role in this.