Problem with amplifying a pressure sensor for measuring blood pressure

I think part of my noise problem is the resistor I'm using now for the instrumentation amplifier.

I don't know the tolerances on the one I have but it's probably not good. The one I was expecting to use, until it was apparent the output of the sensor was 10x what the datasheet said, was very good. Now I'll just need to find one with better tolerances.

I hooked up the resistor with lower tolerances and the signal looks pretty clean with me just blowing pressure in the tube to the sensor.

Words can be such a pain. Lower tolerance is the opposite of +/- smaller percent. Did you really mean that using a lower tolerance resistor made it work when the higher tolerance resistor did not? If so then the resistance you really want is not the one calculated but closer to what the LT resistor actually is.

A 5% tolerance resistor is higher tolerance than a 10% tolerance resistor. The higher the tolerance, the higher the accuracy must be to pass. It's intuitive if you make things to tolerances but otherwise not.

I think that they test the things every so many and if within 10% then that part of the run goes in the 10% bin even as the process is tweaked for higher accuracy. Still there's nothing saying that +/- 10% can't be within 1%, just that's not likely given consistent manufacture.

You're right, I worded it wrong.

I meant the opposite. The good resistor I have is 0.05%, and the other one I'm using I don't know. But I'm guessing probably 5 or 10%.

Your context pointed that way but hey, your thinking counts more than your nomenclature.

The signal is looking better with a better resistor for the instrumentation amplifier. I used a 10.2kOhm/0.1% tolerance. I attached a pdf of the pressure signals again.

The pictures are from 2 different trials.
Oddly, it does seem the device is altering the pressure a bit when it is detecting the systolic & diastolic blood pressure. Cause some noise-like signal appears where those pressures are in each trial.

bpReadings2.pdf (34.7 KB)

I should have thought to mention, add a trim pot to the resistor and you have a chance to tweak the ohms.

Don't trimpots usually have much higher tolerances and produce noisier signals?

Don't trimpots usually have much higher tolerances and produce noisier signals?

Tolerance? They adjust. Noisy? I don't know, mine don't seem to be. Maybe they are -while- they're being turned. Otherwise get audio pots. The whole idea is to be able to tweak the resistance instead of picking the resistor closest to best, since they only make so many values of resistor.

I guess you could also put some fixed-value resistors in parallel to get a between value too. It'd be cheaper and noise-free.

oscarcar:
Don't trimpots usually have much higher tolerances and produce noisier signals?

Trim pots of the 10 and 20 turn variety are often used in precision high quality instrumentation circuits for calibration purposes. One trick is to not let the pot control a larger range then it needs to be, so if you need a +/- 5 percent range of adjustments you might use say a 100 ohm 20 turn trim pot that has it's end terminals wired to 1% 10K fixed resistors. That way any drift in the pot due to tempco would have a much smaller effect in the overall circuit compared to just using a 20K ohm trim pot.

Lefty

Is it really necessary for you to be able to tweek the gain of your instrumentation amplifier? It seems to me you just need to select a precision resistor that gives you the gain you need for the maximum signal level that doesn't exceed your ADC input range. Use the instrumentation amp's gain equation to get the actual gain that results from the value of the gain resistor you actually use.

I imagine you are balancing the transducer at zero pressure in software. I don't think you've talked about how you intend to calibrate it. Maybe the data sheet span voltage is good enough for your application, or maybe the transducer comes with a calibration sheet.

I

t seems to me you just need to select a precision resistor that gives you the gain you need for the maximum signal level that doesn't exceed your ADC input range.

So you can get non-standard value resistors?

GoForSmoke:
I

t seems to me you just need to select a precision resistor that gives you the gain you need for the maximum signal level that doesn't exceed your ADC input range.

So you can get non-standard value resistors?

It seems to me you just need to select a precision resistor that gives you the gain you need for the maximum signal level that doesn't exceed your ADC input range. Use the instrumentation amp's gain equation to get the actual gain that results from the value of the gain resistor you actually use.

Let me state it more simply. Decide what maximum gain you need, based on not exceeding your ADC's input spec. Solve the data sheet equation for the resistor value that will give you that gain. Select the standard value of the precision (1%, 0.1%, 0.05%, etc) resistor that is closest to the calculated value without causing the amplifier to exceed the maximum gain. Then, if you need the actual amplifier gain in your calculations, solve the data sheet equation for gain based on the resistor value you actually use. Is that more clear?

retrolefty:

oscarcar:
Don't trimpots usually have much higher tolerances and produce noisier signals?

Trim pots of the 10 and 20 turn variety are often used in precision high quality instrumentation circuits for calibration purposes. One trick is to not let the pot control a larger range then it needs to be, so if you need a +/- 5 percent range of adjustments you might use say a 100 ohm 20 turn trim pot that has it's end terminals wired to 1% 10K fixed resistors. That way any drift in the pot due to tempco would have a much smaller effect in the overall circuit compared to just using a 20K ohm trim pot.

Lefty

Quite true. Also as microcomputers and SOC's get cheaper and cheaper, nowadays it is becoming quite common to use a fixed resistor and do the calibration in software during final test of the instrument. When it's time for a periodic instrument calibration, if anything changes, the calibration values are simply updated in EEPROM or whatever.

Is that more clear?

I always got it. Calibration/tweaking can be done in software or externally via hardware adjustments. Depends on the application and users requirements. If it's for one's own project then software is cheaper as one has the freedom of recompiling the sketch at any time when external things are changed. If however the project is designed to support changes in sensors or other external changes by users of the project then external calibration might be more desired. At the refinery I worked at we used thousands of sensors wired into central control systems, and we always required that any sensors used had the means to calibrate them externally so that we did not have to have 'custom' loop calibrations for every sensor wired to the central systems.

Different strokes for different folks, there is no single "best method", just what works best for you.

Lefty

retrolefty:

Is that more clear?

I always got it. Calibration/tweaking can be done in software or externally via hardware adjustments. Depends on the application and users requirements. If it's for one's own project then software is cheaper as one has the freedom of recompiling the sketch at any time when external things are changed. If however the project is designed to support changes in sensors or other external changes by users of the project then external calibration might be more desired. At the refinery I worked at we used thousands of sensors wired into central control systems, and we always required that any sensors used had the means to calibrate them externally so that we did not have to have 'custom' loop calibrations for every sensor wired to the central systems.

Different strokes for different folks, there is no single "best method", just what works best for you.

Lefty

When I first read this thread, I wondered if I had the means to calibrate a 0 - 37 kPa transducer. It turns out my smaller Chandler deadweight tester will actually go down that low. I wouldn't be able to match the exact full scale, but 1 psi increments would probably give a good enough calibration.

Didn't realize the thread was continuing. I have been busy reading some books and trying to understand electrical circuits and how to do the current source.

I'm working with an LM317T, but I'm having trouble with the basics. I thought I understood what was going on, but I don't.
V=IR, so since the IC is maintaining 1.25 volts across the "output" and "adjust" pins, I can use a 12.5kOhm resistor.

Oddly, I never get a 1.25V reading across the output and adjust pins. Obviously I'm missing something.
If anyone wants to take a look, this is my understanding of the circuit diagram that I'm trying to duplicate in the breadboard. Attached pdf.

But I am having many difficulties with the basics. The voltages at the output and adjust seem to be zero, with both being about a 0.5 volts less than the input with reference to ground.

I am also having difficulty measuring current. I setup an LED so that I make sure that the current was flowing thru the multimeter, but when I connect the multimeter in series the LED goes off.
My multimeter is very confusing in regard to units of current (uA/mA/A), and seems unstable. But that's probably cause I'm doing something wrong if I can't get current to go through it.

And when I plug the circuit into a higher voltage power source (2.5V, 3.3V) the LED gets brighter, which would seem to indicate more current and thus not working as a constant current source. As I think LEDs get brighter on current and not voltage. When I plug in a 5V power source it starts fading and getting erratic so I guess that voltage is nearly out of range for the LED.

I did manage to get IC fairly warm when I plugged stuff together in certain ways so I assume it does work.

Oh, and yes at some point I can calibrate the sensor. But I still likely need a constant current source so that the calibrations don't change over time.

LM317CurrentSource_bb.pdf (571 KB)

If you look on page 40 of the sensor datasheet, you will see the arrangement they suggest for the constant current supply, using an op amp and three resistors.

Forget the LM317T, its adjustment pin current is around 50uA, so it's completely useless as a 100uA constant current source.

OK, I'll do that. But I'd like to figure out why this circuit is not behaving like I expect. It bugs me that I can't even get 1.25 volts that this IC is supposed to maintain across those 2 pins. I'd like to see if I could get something like 1 mA or something from it.
If someone can tell me that the breadboard looks OK that would be very helpful. At least I would know if I'm understanding the circuit diagrams. Thanks for any help. I think I'll go read some of the general electronics post and see if I can feel comfortable with other circuits and their discussions.

I am using a similar setup. I have a voltage regulator (TI LP2950) going to a resistor (33kohms) and then to the Omron SMPP-O2 sensor pin #6. I set the connections exactly as was posted in this forum. I plan to send Vout(+) and Vout(-) to an opamp, but first I need to get the pressure sensor to work!

I measure Vout(+) and Vout(-) using a voltmeter and I always get a very small reading (~0 mV). This concerns me since oscarcar said that he got -4mv with a null signal. I also have the pressure valve attached to a pressure cuff and when it is inflated the value that I measure with the voltmeter Vout(+) and Vout(-) does not change. I have replaced the first pressure sensor with another. That did not correct the problem. Any advice would be appreciated.