Author Topic: What are typical metal foil strain gauge offset values?  (Read 2841 times)

0 Members and 1 Guest are viewing this topic.

Offline 741Topic starter

  • Frequent Contributor
  • **
  • Posts: 382
  • Country: gb
    • Circuit & PCB Design (small PCB quantities OK)
What are typical metal foil strain gauge offset values?
« on: November 04, 2017, 05:22:35 pm »
I read in an Omega Engineering app note that, whilst one might expect a new gauge to have maybe 1% variation in each 'arm', the 'offset' value typically changes considerably once the gauge is fixed in situ (with adhesive).

Assuming, for the sake of argument, a 1V exitation, what range of "potential divider voltages" would I expect to see in an un-stressed gauge?

If everything was ideal, I'd think expect exactly 0.5V, for instance 350R to +1V, 350R to 0V.

(I'd like to know how much range an offset nulling circuit should expect to cater for, basically).

My guess is I allow a trim of +/- 5mV per half-bridge, thus 10mV 'error' across a full bridge.

Offline Brutte

  • Frequent Contributor
  • **
  • Posts: 614
Re: What are typical metal foil strain gauge offset values?
« Reply #1 on: November 04, 2017, 08:55:08 pm »
With +-1% tolerance you'd get 1% span of the divider, worst case.
So with 1V excitation the output of the divider may vary by +-5mV. From 495mV to 505mV.
With two halfbridges the differential voltage doubles, it is +- 10mV worst case.

Quote
thus 10mV 'error' across a full bridge.
thus +-10mV 'error' across full bridge.
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6927
  • Country: ca
Re: What are typical metal foil strain gauge offset values?
« Reply #2 on: November 04, 2017, 11:07:42 pm »
I found it's not just the resistance variation between gages.
If your gage has load on it at rest (i.e. tare), then your offset is also due to mechanical asymmetry (gage placement, adhesive, metal machining tolerances).

I've had MecE's make strain gages and do a terrible job placing the elements and needed over 20% correction. Your expected span and offset can be way off. You could leave provisions for resistors/jumpers just in case the result is worse. It depends if your strain gage is home made, if it's measuring tension/torque etc.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16272
  • Country: za
Re: What are typical metal foil strain gauge offset values?
« Reply #3 on: November 05, 2017, 07:17:25 am »
I have opened up commercial loadcells, and even there they have had pretty big differences in the trim resistors, and almost all of them require either a calibration cycle or a set of calibration resistors embedded to give a plug in and go unit that relies on a fixed meter.
 

Offline branadic

  • Super Contributor
  • ***
  • Posts: 2380
  • Country: de
  • Sounds like noise
Re: What are typical metal foil strain gauge offset values?
« Reply #4 on: November 05, 2017, 12:50:59 pm »
Quote
I have opened up commercial loadcells, and even there they have had pretty big differences in the trim resistors, and almost all of them require either a calibration cycle or a set of calibration resistors embedded to give a plug in and go unit that relies on a fixed meter.

Well, that's at least one methode.
I do know companies that do the trimming of the load cell just by milling the load cell after everything is assembled to trimm and linearize it. There are other companies depositing the strain gauges directly on the load cell instead of gluing foil based ones (e.g. PVD with an insulator first), which always results in an offset beside the initial offset of the strain gauge itself. The strain gauge can then be trimmed by laser, but there is no drift caused by relaxing glue over time.
Another methode is the digital way, which results in a loss of dynamic range.

-branadic-
Computers exist to solve problems that we wouldn't have without them. AI exists to answer questions, we wouldn't ask without it.
 
The following users thanked this post: zhtoor

Offline 741Topic starter

  • Frequent Contributor
  • **
  • Posts: 382
  • Country: gb
    • Circuit & PCB Design (small PCB quantities OK)
Re: What are typical metal foil strain gauge offset values?
« Reply #5 on: November 05, 2017, 05:49:40 pm »
Thanks for your responses.


Brutte: My real concern is that the '1%' figure seems to be not something to rely on once the gauge is affixed in place. That is, this figure is perhaps only of incidental use, as the in-place offset can be more.


floobydust:, Thanks, looks like I have to allow for far larger offserts than I guessed.

I'd be trimming this 'digitally', perhaps with a digital 'pot' across 2 of the resistors. If I knew the trim would be very small, and within fairly certain limits, I could optimise the values to get good resolution.

Of course, I still have temperature effects at the gauge and throughout the measurement system...
« Last Edit: November 05, 2017, 05:57:16 pm by 741 »
 

Offline Vtile

  • Super Contributor
  • ***
  • Posts: 1144
  • Country: fi
  • Ingineer
Re: What are typical metal foil strain gauge offset values?
« Reply #6 on: November 05, 2017, 09:04:01 pm »
Also take note of the electrical loading coming from the measuring device. It can be surprisingly big depending the accuracy needed** and most appnotes I have read (not many) do consider it "immune to loading effects" as it is "balanced" circuit, the reason I say this is because they do not talk about it.

** ..and how far is the operation point from the true zero/balance point of the circuit.
 

Online wraper

  • Supporter
  • ****
  • Posts: 16803
  • Country: lv
Re: What are typical metal foil strain gauge offset values?
« Reply #7 on: November 05, 2017, 09:53:14 pm »
I'd be trimming this 'digitally', perhaps with a digital 'pot' across 2 of the resistors. If I knew the trim would be very small, and within fairly certain limits, I could optimise the values to get good resolution.
Why would you do this? Unless you need analog output to connect some external device which cannot make adjustment to the reading y itself, this is pointless waste of parts / money and compromised linearity. Normally you should just digitize it, and remove offset in software. Almost all scales take zero reading on power on.
 

Offline 741Topic starter

  • Frequent Contributor
  • **
  • Posts: 382
  • Country: gb
    • Circuit & PCB Design (small PCB quantities OK)
Re: What are typical metal foil strain gauge offset values?
« Reply #8 on: November 06, 2017, 06:26:14 pm »
Hi,

I do not want to waste dynamic range in the amplifier by amplifying the offset.

The signal is small, the gain is large.

If I had arbitrary dynamic range, then sure, I'd simply subtarct the offset in software. In fact - that was my original plan.

Now, having got an idea of just how large the offsets can be, I feel it is safer to allow this option. say there is 50% offset - I'd really struggle to accomodate that without true hardware offsetting.

It 'costs' a couple of digtal pots: these can be small & cheap.

Online wraper

  • Supporter
  • ****
  • Posts: 16803
  • Country: lv
Re: What are typical metal foil strain gauge offset values?
« Reply #9 on: November 08, 2017, 12:26:12 pm »
Hi,

I do not want to waste dynamic range in the amplifier by amplifying the offset.

The signal is small, the gain is large.
Whaat? Offset of the cell itself is maximum a few % of the range. Then there will be some added load on it once you put it in your system. In any case you cannot get over 100% of the range without overloading the load cell.
Quote
say there is 50% offset
If there is positive 50% offset, you cannot add more than additional 50% of the rated max load without going past maximum load cell spec. If there is negative offset, you are doing something strange.
EDIT: I thought he just using off the shelf load cell, not attaching separate strain gauges.
« Last Edit: November 08, 2017, 09:54:13 pm by wraper »
 

Offline voltsandjolts

  • Supporter
  • ****
  • Posts: 2282
  • Country: gb
Re: What are typical metal foil strain gauge offset values?
« Reply #10 on: November 08, 2017, 01:25:02 pm »
I do not want to waste dynamic range in the amplifier by amplifying the offset.

Yes, exactly. When the measurand is small compared to the offset, the offset needs to be removed before applying gain.

There are a lot of options but I find the TI PGA308 useful with digitally programmable gain and offset.

 

Online wraper

  • Supporter
  • ****
  • Posts: 16803
  • Country: lv
Re: What are typical metal foil strain gauge offset values?
« Reply #11 on: November 08, 2017, 02:37:08 pm »
I do not want to waste dynamic range in the amplifier by amplifying the offset.

Yes, exactly. When the measurand is small compared to the offset, the offset needs to be removed before applying gain.

There are a lot of options but I find the TI PGA308 useful with digitally programmable gain and offset.
Adjusting offset in analog has barely any effect on dynamic range unless your ADC does not have enough effective resolution to do the task.
EDIT: I thought he just using off the shelf load cell, not attaching separate strain gauges.
« Last Edit: November 08, 2017, 09:53:31 pm by wraper »
 

Offline voltsandjolts

  • Supporter
  • ****
  • Posts: 2282
  • Country: gb
Re: What are typical metal foil strain gauge offset values?
« Reply #12 on: November 08, 2017, 02:41:59 pm »
 :palm:

It does if you want x1000 gain before the ADC
 

Offline Brutte

  • Frequent Contributor
  • **
  • Posts: 614
Re: What are typical metal foil strain gauge offset values?
« Reply #13 on: November 08, 2017, 07:12:11 pm »
Of course you need an offset cancellation in analog domain.
Metal foil strain gauges do not change the resistance anywhere near 1e-2 in their operating range. Closer to 1e-3.
Also the arrangement matters as this can be full bridge with one active element or full bridge with four active elements. So if the offset is comparable to the operating range of the output signal then ok, you can calibrate offset in digital domain. But if the signal is an order of magnitude smaller than offset then that is a different story.

 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 6927
  • Country: ca
Re: What are typical metal foil strain gauge offset values?
« Reply #14 on: November 10, 2017, 04:09:33 am »
I don't like padding resistance on the gages to correct offset as this also changes span- I can't recalibrate a load cell (span) everytime the offset is adjusted, not practical.

What's worked for me is a front-end pre-amp x100 gain (instrumentation amp) with DC offset null. This then feeds the A/D, now that most/all of the offset is gone. You can also zero what remains in software.
If the pre-amp's range can't adjust the offset, I know the bridge is crap and needs to be fixed first.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf