Trouble adjusting OP offset voltage.

[LEECH666]

Hi,

for my dummy load project I wired up a non inverting summing amplifier on my breadboard according to this circuit diagram (my own).



I wired it up slighly different, but quite similar:
- Only 1 pot for adjusting the gain (RV5)
- Only two 10K summing resistors (R4, R5) on the input pin 3



When I short the input pins 2 and 3 together I get around +11mV offset on the output pin 6. The OP I am using is a TS271 from ST (TLC271 equivalent).

When I turn the potentiometer RV4, nothing happens. I am not sure why.
I also tried a smaller single turn 5K pot, but no success.
Shouldn't I be able to adjust to wind this voltage down to zero volts this way?
Is it possible that I either need a negative voltage supply or a rail-2-rail OP to go down to zero volts?

Any help is appreciated.

Cheers,
Florian

[w2aew]

Two things:
1) Looks like you omitted the Rset resistor on pin 8 to program the Iset current.  You need this to make the opamp work.
2) Its not clear what the load draws, that will determine how close the output will be able to swing to the bottom rail.

[LEECH666]

Thank you for your reply. I guess I should have chosen a more basic OP as this bias selection stuff is a bit atop of my head.
I also guess I should have taken a closer look at the datasheet, the offset adjust for single supply seems to be slightly different.

http://www.ti.com/lit/ds/symlink/tlc271.pdf
See page 53, Figure 105 (a).

Not sure which bias mode I want to use.
I guess I will just go with high bias for now and hence connect pin 8 to ground. (as shown in figure 1 on page 3).

Thanks for the help w2aew, I will try out your suggestions. If all fails I will try a TL071 instead of the TLC271.

Back to the bench. 

Cheers,
Florian

[w2aew]

Not sure which bias mode I want to use.
I guess I will just go with high bias for now and hence connect pin 8 to ground. (as shown in figure 1 on page 3).

Thanks for the help w2aew, I will try out your suggestions. If all fails I will try a TL071 instead of the TLC271.

You can easily try the different bias selections.

The 071 isn't well suited for single supply operation.  It's output can't swing to the rails. Stick with the 271 for now, it should work well.

[LEECH666]

Man I feel a bit stupid for looking at the TI datasheet instead of the ST one. Apparently the two types (TLC271C and TS271C) are not 100% equivalent.

Anyway I managed to adjust the offset to 1.2mV (according to my Fluke 87-IV) with the diagram shown in figure 105 (a) from the TI datasheet.

However my scope is showing me different values (10.4mV - see pictures):





I enabled 4 sample averaging and HF noise rejection in the trigger menu because there seems to be some kind of noise / ringing on the signal. Probably related to the ground leads of my probes.

With no offset compensation potentiometer connected the offset voltage is around 55mV.

After looking at the datasheet of the TS271 I am confused about the value I should set ISET to. It says that it adjusts the speed vs power consumption of the chip. What do they mean with speed? I guess they mean slew rate with that.

I don't really care about power consumption for now so I guess what I want is speed. ISET = 130µA is the highest speed they recorded electrical characteristics for, so I guess I'll just go with that.  180K from pin 8 to GND seems about right for 130µA (see figure 6 / page 5 - http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00000494.pdf)

Ah well it's bedtime in Germany now ... 

Thanks for your help again w2aew.

Cheers,
Florian

[w2aew]

10mV might be as close as the output can swing to ground when ground is used as the negative rail.

[LEECH666]

I guess I could try my luck with something like this to split my 12V into ±6V:



R1 = 4.7K-1/2W
R2 = 4.7K-1/2W
C1 = 100nF-63V
C2 = 220µF-25V
C3 = 220µF-25V
C4 = 100nF-63V
C5 = 100nF-63V
Q1 = BD437
Q2 = BD438
IC1 = LM358

I'm just gonna use what I have at hands tho (TL071, BC338, BC328, 100µF) ...

Prototyped this in Blackboard v1.0 (*click*).



Not quite Sijosae, but close.

Haven't tried the 180K to ground on pin 8 of the TS271C yet. Guess that's what I'm going to do now.

//EDIT, okay tried the 180K, now I can adjust the offset between 8mV and 8V, but not infinitely variable, just two modes: 8mV or 8V. Pshhhhh .... not sure what's going on here, or if this is the way it's supposed to be.

Cheers,
Florian

[LEECH666]

Virtual ground circuit works. I will try it on the summing amp tomorrow (probably).

Cheers,
Florian

[say]

Hi Leech, thanx for directing me here. Although I am not an expert in this stuff but seeing your first circuit at the top, it appears the gain is set too high for the summer circuit. As far as I can tell the voltage at NI terminal is an average of the input signal. Thus if you have two inputs you should have a gain of two, for four inputs the gain should be 4. That will give you a true output equal to the sum of inputs and any error there could be adjusted with a single pot of say 50k ohm (I suppose you won't get 40k pot).

Cheers

[LEECH666]

Grrr. 

Dodgy / defektive OP amp discovered. Was about to post a rant post but then I switched the OP for a previously unused one. I was seriously starting to doubt my electronics education, not being able to wire up a simple circuit like a summing amp. *sigh*

[LEECH666]

And I just realized that I'll need a non inverting summing amp to make it work of a single supply. I thought the circuit was going bonkers again ... but yeah I am just a DERP! For some reason I wired up a normal summing amp. *sigh* I should update my shematics more often.