I cannot use micro-controllers for the project, is there any other method?

 

how do you expect people to answer with absolutely nothing to go on?

 

I cannot use micro-controllers for the project, is there any other method?

Yes there is.

 

Ok the basic problem I have at hand is to get some temperature sensors that output a voltage of some factor of temperature. I want to add these voltage signals to produce one sum signal. I have like 5 or 6 of these temperature inputs. How can this be done without using Micro-Controller? I already have the sensors complete.

 

Ok the basic problem I have at hand is to get some temperature sensors that output a voltage of some factor of temperature. I want to add these voltage signals to produce one sum signal. I have like 5 or 6 of these temperature inputs. How can this be done without using Micro-Controller? I already have the sensors complete.

It's too bad you can't use a microcontroller, because that is ideal for this application. You would have the most flexibility for sensor types and could handle nonlinear responses and complicated mathematics etc.

You may need to provide more information on how you want to measure temperature. Are you measuring local air temperature, or do you need to have remotely located temperature probes? Also, what is the temperature range?

However, in principle this could be quite simple. In another thread, someone recommended the LM35 temperature sensor which provides a linear voltage output for temperature in Celsius degrees. These linear outputs can be added with an simple OPAMP adder circuit. If you can use the LM35 (or emulate the linear response with your sensor circuit) your solution is very simple - almost trivial.

I suspect that the LM35 might not meet your needs, in which case you would have to linearize the nonlinear response of whatever temperature sensor you chose. This could be simple if the temperature range in small, but it could be very challenging if the range is large.

Can you provide more information on the sensors you have, type and whether you have already linearized the response?

 

Rather than summing the temperatures, wouldn't it be more meaningful to average them?

 

I'm not up on this stuff. I used a multiplier chip years ago, but can't remember which one. I think it might have been an Analog Devices chip. I recommend looking at the Analog Devices web site and read through the data sheets. If I remember right, these parts aren't cheap. Nowadays, you would be much better off with a digital approach. Maybe the only advantage of analog parts (for this type of application) could be high bandwidth. Since we don't know your application, it's hard to say much more.

Hi Steve:

Analog Devices seems to be the main player in the LogAmp business. We used a lot of their stuff in instrumentation i the UCLA plasma lab. Any complex math function can be done with them, as you mention. This is pretty specialized knowledge, and not too many players in the field. Sort of a shame too.....there's a mentality out there that says, "If it can be done with a microprocessor, why learn analog math?" Don't even get me started on this one!

Eric

 

The range for the project is not very large (15 - 35 C at most). The probes take the temperatures of different objects. I looked at the OPAMP adding circuit and it uses superposition of the inverted AMP configuration. Can the adding circuit be constructed using the non-inverted configuration?

 

The range for the project is not very large (15 - 35 C at most). The probes take the temperatures of different objects.

Can you say what type of probe and what accuracy you need. I once linearized a thermister over this exact temperature range. I used a 10K thermistor with a resistor in parallel, then another in series with that, and then another in parallel with that combination. Then I drove the network with a current source. I can't remember for sure, but I think I got better than 1 degree linearization over that temperature range.

I looked at the OPAMP adding circuit and it uses superposition of the inverted AMP configuration. Can the adding circuit be constructed using the non-inverted configuration?

The simplest thing is just to use an inverting adder and then an inverting amplifier after that. This gives good performance and a great deal of flexibility on the gains. OPAMPS can be purchased in dual or quad packages, so this is straightforward.

 

I have a basic electronics question. If you have 3 voltage signals from 3 different wires and you combine the wires. Do you get the sum?

Also, I am using the LM35, is there anything wrong with that?

 

Three wires? No you get a partial short circuit. Try this link - http://focus.ti.com/lit/an/slaa068a/slaa068a.pdf.

Whoops, the link doesn't get to an article. Here is one that shows how to make a resistive mixer - http://devices.sapp.org/circuit/mixer/. Just reinvert afterward.

 

I have a basic electronics question. If you have 3 voltage signals from 3 different wires and you combine the wires. Do you get the sum?

It depends on what you mean by combine. Based on your previous information, your 3 voltage signals need to be referenced to the same ground level and then you input them into the separate inputs of the summing amplifier that has the same voltage ground. I would recommend that you post your schematics, and this will make it easy for other people here to comment on. You should provide details on the distance between the sensors. If you have remotely placed sensors or circuits, there are special considerations for grounding and noise.

Also, I am using the LM35, is there anything wrong with that?

I see nothing wrong with this. If you feel these do what you want, they should be OK. The good thing is that these provide linear output, so you can directly add the voltages as you want.