I feel it is to do with the capacitor destabilising the amplifier inside the transducer as OBW0549 has said. (...) I tested a sample of 10 similar transducers today. Each one has this d.c. offset voltage... But not all the same value - most aren't out by enough to affect the performance of the system, but some clearly are. Does anyone know if the offset is likely to change over time? I can accept some margin of error as long as it will perform consistently. i.e. if there is some offset will it get worse?

I'd say that unless you've conclusively established the cause of this offset (not just suspecting it) and found a cure that's 100% effective in every case, it's impossible to know if the offset is going to change with time or be consistent. If it were me, I wouldn't count on consistency and would redouble my efforts to find the actual cause of the offset-- your sensors simply should not be behaving that way.

Frankly, at this point I'm at a loss as to what the problem could be.

 

I agree, we could dampen it mechanically but the operating pressure is up to 150 bar and I believe the components aren't 'budget friendly'!

Dean

... understood.
If nothing economical seems to fix the problem, take a look at this item. It seems to comply with the physical demands which you describe. As you say, it's not available at the local store.
https://store.livhaven.com/378190-h...tD1diJwx-oQZ8TXe0aq6_02K3FHnVfIaAqGIEALw_wcB#

... the key feature would be the line from the datasheet:
• Extremely quick shock response

 

The manufacturer has confirmed that the capacitive load limit of the sensor is 100 nF so my capacitor has to go! I thought about this type of circuit to try and fix it. Its a low pass filter with a voltage follower with a capacitor based on the maximum load of 100 nF, and I will try some different values of resistor from about 1M to 5M and see which gives the best filtering for the application (0.3 - 1 Hz) - response time doesn't have to be fast.

Since the resistance is huge in comparison to the 12k input impedance of the controller I hope to be able to use a voltage follower with a high impedance to preserve the signal. I have no idea what component to use for the voltage follower though so any guidance would be massively appreciated.


Dean

 

Because of the high source R to OpAmp you want one with low
bias current so that its bias I does not create a lot of offset. Also
since you are in single supply application a rail to rail input and output
OpAmp would make sense. OpAmpBW in the .1 - 5 Mhz range would
do. Might as well make it a low power OpAmp as you have no need
for any significant speed.

If you go to TI or others they have a tool to narrow your choices.

http://www.ti.com/amplifier-circuit/op-amps/products.html#p23typ=1;3&~p78=In;Out&p1261max=10;180&p1261min=0.9;10&!p1498=HiRel Enhanced Product;Space;Automotive

Check boxes rail to rail in and out, set GBW for 1 Mhz, power for +5 and that
will narrow your selection.

Regards, Dana.

 

Check boxes rail to rail in and out, set GBW for 1 Mhz, power for +5 and that will narrow your selection.

5 VDC for power would be inadequate, as his signal goes from 0 to 10 V.

 

Good point OBW0549, I need to keep my readers on at all times.

Potatohead modify your selection criteria to 12V or greater, or run it off 24 and
add some margin and screen for that.

But this raises the question what is PLC input range ? You may need to use a
voltage divider to conform to its input limitations.


Regards, Dana.

 

From Post #1

I tried a couple of different transducers of similar spec but from different manufacturers and this problem did not exist.

So why not just use those transducers instead of all this fooling around with filters and regulators?
SG

 

From Post #1

So why not just use those transducers instead of all this fooling around with filters and regulators?
SG

Good question

I wanted to understand why the problem is occurring and I have found out - the capacitive load is too high. Yes the problem didn't exist on those transducers, but when I tested a whole batch of the original ones, I found there were some with a really bad offset and some were barley noticeable.

What if i got lucky and the other transducers I tested were just on the good end of their tolerance (I had a sample batch qty of 1 for each type) for example, and I replaced a large amount of them at significant cost (each one was twice the price in fact) to start noticing that those turn out to exhibit the same problem... In fact it just seems like its bad practice to hang a large capacitor off of a transducer in general and I'm glad i found out.

The spikes still need filtering though so the best solution to me is a filter that is sympathetic to the workings of the sensor, i.e. less capacitive, which is where fooling around with the filters and regulators comes in.

Dean

 

which is where fooling around with the filters and regulators comes in.

Understand. In that case use a buffered filter.
SG

 

An LM358 might not be best choice, you need to examine CMR at the input
to see if it will fit design. A RRIO OpAmp may be a better choice. This will
depend on your OpAmp power supply Voltage.

https://www.analog.com/media/en/training-seminars/tutorials/MT-041.pdf


Regards, Dana.