Hi,

I'm using a DT85 for reading HIC stress cells containing 12 strain gauges. Gauges can be temperature compensated afterwards with a separate RTD instead of a dummy gauge.

Two common grounds are available too, thus making three-wire quarter-bridge reading method possible. I'm wondering using just a one single precision bridge completion resistor to read all gauge channels.

Should this be possible without sacrifice of accuracy? Inactive channels of DT should be totally disconnected or Hi-Z mode hence not affecting the bridge completion resistor?

Good morning bladE_666,

Two common grounds??? Sorry but I don't understand. Yes you can. If you look at the DT500 range manual on page 19 configuration 14 will be identical for DT80 range.

Cheers,
Roger

Dear Roger.

I appreciate your fast reply to this topic, it seems that coffee tastes good at there. I should have read the manual more carefully, but a high multiprocessing workload can be very tricky sometimes.

Common grounds...probably that wasn't the keyword (my bad; let's try something else then). I mean common negative and return signals from the strain gauge cell. This mean that there is no individual return or ground signals for strain gauges of the cell.

Config. 14 was pretty straightforward. My configuration is a slightly different. I think it will do, but I'm not 100% sure will it cause some interference or deviation sooner or later?

Good afternoon bladE_666,

This wiring configuration is not in DT80 user manual but it can be done on DT80. Is there two separate wires inside the device (as shown) or is there one sample point? One sample point would be preferable as you won't then have a double path.

While 1.2 Ohms may not seem much of an offset it does represent a zero shift of 4789 micro strain on your 120 Ohm strain gauges so keep the wires inside the device and at the dataTaker as short as possible.

Regards,
Roger

There is a two separated wires inside of the cell if I understand your question right? I have assumed that two commons are needed for lead resistance compensation if excitation and return lead resistances are identical? So I'm wondering about huge zero shift of 4789 strain.

Excitation signals to all gauges are individual, I have a little problems to understand what do you mean with one sample point? To make things even worst the cell is installed into a borehole depth of 10 - 30 meters. Thus making measurements at 'close contact' impossible. However wires inside of the logger box are as short as possible.

I attached to schematic which I have created own my own, since I have difficulties to get it from manufacturer. It should be ok and maybe it will help all of us: http://smoy.pp.fi/temp/Upload/HIC_sch.pdf

I have not yet familiarized myself too specific how to calculate to PPM and strains, but is the connection below correct to convert voltage difference into PPM? I was able to get 5025 PPM and 4718 micro strains while GF is 2.088.

Good afternoon bladE_666,

Ok, Now this is different to what we have been discussing. U1 is reading the output of a full bridge output and U2 is reading a quarter bridge of the same bridge...

I'll need to talk to the hardware engineer about this and get back to you.

As for how the dataTaker half bridge works. We supply a current to the * terminal the current flows through R1 and Rc and back to the #. We then measure the actual current and calculate the Voltage across R1 and Rc.

While this is happening we have hardware that measures the lead resistance between the - and # and subtracts this value from the reading. (Notice that this includes the value of Rc and allows us to measure the change in R1 as a ratio)

In your previous diagram you show the connection to the - terminals as branching in the device. This means you have two paths in parallel which will reduce the resistance in that part of the circuit( 1/Rt = 1/R1 + 1/R2...+ 1/Rn).

As change in resistance across a strain gauge (DR) = Gauge factor (GF) X Gauge resistance (GR) X strain reading (e), then ue=DR /(GR x GF)
So if the two wires in parallel reduce the resistance by 1 Ohm then
e = 1/(120 x 2.088) = 0.00399.1 = 3991 micro strain.

This is why the question on exactly how it is wired internally. But then it is an offset which you can zero out...

Cheers,
Roger

Good morning bladE_666,

The wiring above doesn't tally with the resistances reported on the wiring sheet. If we measure the resistance between points 2 & 3 we have two resistors in series on each arm so the resistance will be:

1/Rt =1/(120+120) + 1/(120+120) = 1/0.00833

There for the resistance between 2 & 3 is 120 Ohms. If we measure resistance between points 1 & 2 then we get:

1/Rt= 1/120 + 1/(120 + 120 + 120) = 0.0083 + 0.00277 = 0.01111

There fore the resistance between 1 & 2 is 90 Ohms. We really need to work out exactly how this device is connected internally.

Cheers,
Roger

Hello again

Above screen shot is misleading a little bit. It is not directly related to the stress cell. I was wondering PPM calculations at the same time which leads sprawling thread indeed.

Stress cell is connected to the logger with modified configuration 14. It seems to work pretty nice

I have started my summer holidays, but I'm still unable to forget this project for a while. I have one question, if I could to get some verification? Raw data values of PPM are converted to strains with using formula (2*PPM)/GF, is this correct?

I found some document from DT web site, which offer a different formula of (1/GF)*PPM. This half the results when compared to the first equation.

Another issue which I just recovered yesterday is regarding to download the CSV data from logger. Maybe I explain a little more. Purpose of this project was made to rugged portable data logging unit to read multiple stress cells one at the time. For this interactive functions are created so user can select which stress cell is being read.

This is useful since cell specific gauge factors can be used as well the logged data is columned to each stress cell. With enough columns or channels defined downloaded CSV file get just corrupted without any data. Did I have miss something?

Good morning bladE_666,

1. Strain calculations.
   First off you are reading DT500 technical reference manual. Not all of it applies to the DT8x so use with care. (Note the manual is not very clear but is for two active elements). Do you have one or two active elements?

   The formula is as stated: ue = PPM x (2/ (N x GF))
   Where
   ue is microstrain reading
   PPM is the DT80 reading
   N is the number of active elements
   GF is the gauge factor.

2. CSV data.
   I would need to see the data, method of down loading, servicedata report program code etc..

Can you please email this to me and I'll take a look at is all.

Cheers,
Roger

Good morning...

I was thinkin' something like that, but wasn't sure. Thanks for your clarification. There is only one active gauge at the time, so ue = (2 PPM)/(GF N) or same equation with simplificated of yours "ue = PPM x (2/ (N x GF))" will do then, chool.

Not so cool was the error message below. My email address should be fine, but no. I tried to send the mail directly to you.

With best regards,
bladE_666

Good morning Klaus,

I've looked at your data and it doesn't make any sense to me. I think you really need to talk to the manufacturer and find our exactly what the internal wiring is and exactly how it is supposed to be powered and read.

Cheers,
Roger