Operational Trans-Resistance Amplifier

Thread Starter

OTRA868

Joined Jan 30, 2021
3
Hi everyone,

I am interested in the Operational Trans-Resistance Amplifier (OTRA). I understand this topology can be used to realize any transfer functions.
I would like to design a High Pass Filter using this method. Anyone familiar with this?
 

Papabravo

Joined Feb 24, 2006
21,158
I'm not aware of any compelling reason to do this. OTRA filter blocks cannot be cascaded to make higher order filters so their performance would be limited to a single stage. Capacitors and inductors work in filters because of their effect on voltage, not current. I'm note even sure there is a filter theory that works with a current to voltage converter. Do you know something that I missed?
 

Thread Starter

OTRA868

Joined Jan 30, 2021
3
I'm not aware of any compelling reason to do this. OTRA filter blocks cannot be cascaded to make higher order filters so their performance would be limited to a single stage. Capacitors and inductors work in filters because of their effect on voltage, not current. I'm note even sure there is a filter theory that works with a current to voltage converter. Do you know something that I missed?
It's for a school project I'm currently working on, from research a OTRA can realize any nth order transfer function by selecting admittance branches. I would like some help in calculating the admittance values for a High Pass filter using the OTRA.
 

Papabravo

Joined Feb 24, 2006
21,158
It's for a school project I'm currently working on, from research a OTRA can realize any nth order transfer function by selecting admittance branches. I would like some help in calculating the admittance values for a High Pass filter using the OTRA.
I'll take your word for all that. Might I suggest you invest in the free simulator LTspice from Analog Devices. There are two main reasons:
  1. It has a very easy and intuitive schematic drawing capability, that will allow you to share and publish your work to a project group for example.
  2. Simulation makes it easy to confirm and inform your methods.
Would you mind showing me an example of "selecting admittance branches". That term does not evoke any meaningful information for me.
BTW simulating an arbitrary transfer function is not exactly a big deal.
 

MrAl

Joined Jun 17, 2014
11,389
I'm not aware of any compelling reason to do this. OTRA filter blocks cannot be cascaded to make higher order filters so their performance would be limited to a single stage. Capacitors and inductors work in filters because of their effect on voltage, not current. I'm note even sure there is a filter theory that works with a current to voltage converter. Do you know something that I missed?
Not sure what you mean here. The way i understand them is they can be used similar to any op amp. The only difference is conceptual, but then again we can think of an ordinary op amp in the same way where we analyze more from a current standpoint than a voltage standpoint (ie the current from input to output is equal to the current from the very input to the op amp input).
Capacitors work on voltage differences but if you have a voltage output and more or less current input and the input is at virtual ground, the capacitor is going to pump current into the input terminal due ot the change in voltage on the output, just like a regular op amp except now the input is at virtual ground no matter what the output is (ie it does not depend on the topology anymore to maintain a virtual ground). So the only difference i think is what forces the virtual ground, the output or the op amp internal design. This makes me think we can use one almost the same way we use any other op amp, which of course means we should not have much trouble cascading stages for a more complex response.
What method he wants to use here though i dont know...choose admittance branches...he will have to elaborate. Maybe he means choose branches to add to the circuit based on individual admittance and known behavior of such branches, which is a very simple technique. Cant be sure until he replies though.
 

Thread Starter

OTRA868

Joined Jan 30, 2021
3
Sorry for the confusion, What I would like to do is convert the transfer function for a let's say 2nd order high pass active filter to one to fit the OTRA topology. I don't think I can post the paper on it here but it is available online "Realization of n-th Order Voltage Transfer Function Using a Single Operational Transresistance Amplifier" In this paper they used a method to realise the transfer function and get RC values. At the moment I need help with the math to convert a 2nd order high pass filter to to OTRA topology and get the values. Hope this helps explain the situation.
 

Papabravo

Joined Feb 24, 2006
21,158
Not sure what you mean here. The way i understand them is they can be used similar to any op amp. The only difference is conceptual, but then again we can think of an ordinary op amp in the same way where we analyze more from a current standpoint than a voltage standpoint (ie the current from input to output is equal to the current from the very input to the op amp input).
Capacitors work on voltage differences but if you have a voltage output and more or less current input and the input is at virtual ground, the capacitor is going to pump current into the input terminal due ot the change in voltage on the output, just like a regular op amp except now the input is at virtual ground no matter what the output is (ie it does not depend on the topology anymore to maintain a virtual ground). So the only difference i think is what forces the virtual ground, the output or the op amp internal design. This makes me think we can use one almost the same way we use any other op amp, which of course means we should not have much trouble cascading stages for a more complex response.
What method he wants to use here though i dont know...choose admittance branches...he will have to elaborate. Maybe he means choose branches to add to the circuit based on individual admittance and known behavior of such branches, which is a very simple technique. Cant be sure until he replies though.
The OTRA takes a current and puts a voltage out. So you can't directly take that voltage output and run it to a current input. Not directly at least.
 

Papabravo

Joined Feb 24, 2006
21,158
So they are talking about doing this on a silicon die. Although it would be nice to have a mini foundry in the basement, that is not likely to happen any time soon. If you can't get your hands on a finished part, how exactly are you going to experiment with this concept.

EDIT: I see a vague suggestion that you can synthesize one with a pair of AD844's. The picture is kinda vague on how the two are actually wired together. The !@#$%^&*() journal authors NEVER put anything useful like...oh....PIN NUMBERS, on a schematic. That would be just oh too...too...much...ah... trouble. Maybe somebody with more analog chops than I've got can give the TS a hand.
 
Last edited:

MrAl

Joined Jun 17, 2014
11,389
The OTRA takes a current and puts a voltage out. So you can't directly take that voltage output and run it to a current input. Not directly at least.
Yeah but you almost never do that anyway, except in the case of a voltage follower you always use resistor(s) between output and input, and that allows you to calculate the voltage input op amp the same way you calculate the current input op amp, because they both can be viewed as having current inputs. The difference there is that the voltage input op amp has the currents go AROUND the op amp while the current input op amp has the currents going INTO the op amp also to going around.
When the voltage input op amp is viewed as a current operating device, we can calculate the current going into the node that forms the inverting terminal and equating that to zero, just like the current input op amp. So the difference is subtle but very nearly the same anyway.

The term "transresistance" however makes me think of a resistance in and a resistance out.
He seems to be suggesting also that we only need one op amp to perform a multipole filter, so i have to wonder if this is just because we can add any number of caps and resistors to get there even without an op amp or there is something extra special about this particular type of op amp.
I'll have to look around the web.
 
Top