Inverting and non-inverting op-amps

I am wondering if anyone know any good webpages that describes different op couplings for creating +/- x V from x V. Or have any advice on the following problem;

I am having a bit of a problem getting a circuit working, need to get up to 5V totally between E+ and E- in the following circuit with a Vref of +2,5V. I use rail-to-rail op´s, feeding +/-5V but no success getting any more than 4,6V.

I thought that if I had a reference of say, 2,5V, in on the top op-amp non-inverting op-amp, I would get a potential difference of 5V between E+ and E-, with E- equal to -2,5V.

http://www.student.itn.liu.se/~peter619/Schema_2505.doc

Could a solution be to feed the system with +/-7V instead of +/-5V as I now do? As it works then, but I do not want to use any thing else than +/-5V..

Any ideas? Thankful for any expert advice to a novice

 

Your circuit is a bit lacking in detail to sort out.

Could you add to it a bit. eg. Is N4 a dual op-amp? I assume the - and + on its RHS are inputs. Are the + on the LHS the outputs? What is N5 and what is it supposed to do? I assume J01 to J04 is a 4way selector switch. Right?
What is the value of zener V06?

If you could indicate how it is supposed to work it would help.

 

Going by what you say without spending a lot of time trying to decipher your circuit, I can say this: rail-to-rail opamps are not, they are actually close to rail-to-rail. Look carefully at your op amp datasheet and look for the output voltage spec. Given that your are within 200mV of the output per side, I would say that is correct. You will need to go to a higher V+ and V- to obtain your desired output (gotta have headroom).

 

Hi! I am really sorry, N4 is a dual op-amp, AD648 actually. N5 is an reference voltage circuit, AD548 with J01 to J04 as voltage dividers for getting reference voltages to N4a´s opamp´s positive input in ranges of 7,5V (giving +/-7,5V, total of 15V), 2,5, 1 and 0V. The zenerdiode V06 is an BZX55/C3V9, with a spec. zenerdiode range of 3,7 to 4,1V. As it is now, N5 is not really connected, I will only have a reference of up to 2,8V.

The circuit is simply supposed to take in a reference voltage and reflect it so that E+ and E- is giving total of say 15V (for input of 7,5V). E+ and E- is then connected to supply an ordinary Wheastonbridge so that S+ and S- can be analyzed

The circuit has originally been designed for a supply of +/-15V and I have tried to lower the supply down to +/- 8V where it still works for reference voltages up to 2,6V, giving out app. 5,2V between E+ and E-. The problem is that I have been given +/-5V only.

I guess the problem here is that both the transistors V01 (2N2222), V03 (2N3904) and V05 (2N2907) togehter with the AD648´s opamps cannot handle low supplies (only got out something like 3,68V between E+ and E- at the most att Vsupply +/-5V). Changed the AD648 to an dual TS462CN, which gave me an output of 4,2-4,6V between E+ and E-..a bit better but still not quite enough (need app.5,2V).

Could a solution be to build a step-up/bost converter that changes 5V to +/-8V???

As you said nomurphy, gotta have headroom..


Please let me know if any details this time is unclear

 

Hi Beagles,
I've checked the AD648 and it has not rail to rail output. The graph was a bit iffy but I think it correspond with your measured output.

Looking at the schematic, you need at least 1.4V extra on the positive side and 0.7V extra on the negative side, on top of the +/-2.6V. So the output you need is min. +4V and -3.5V.

You must use an opamp with rail to rail output, like LM6132 from National. This is just an example. With this you should get the +/- 2.6V output with a supply of +/-5V.

TOK