TDS measurement of the boiler

[gladkenn]

Hey guys,

A little help in here.
I have read in a literature from spirax sarco that the conductivity of boiler water is dependent upon its temperature. I was so struck by this statement because our current practice for measuring the TDS of the boiler water is to get the sample from the boiler and let it cool down in an enclosed bottle. After the temperature has cooled down to its ambient temperature, we measure its conductivity. To convert conductivity into TDS of water, we just multiply it by 0.7 (TDS = 0.7 x conductivity).
If we follow a statement from spirax sarco literature, it seems that we are doing a very inaccurate measurement of the true value of the TDS inside the boiler. We are actually understated the TDS of the boiler.
Any suggestions regarding this?

 

[metengr]

The measurement of TDS using the conductivity method requires the boiler feedwater sample to be near 25 deg C, and pH neutral. The conductivity is affected by pH. I would suppose you are operating at a pH range between 8.5-9. In this case, you need to add acetic acid to reduce the pH to 7 prior to measuring conductivity.

 

[Keef]

Gladdkenn,

The conductivity level is dependant on temperature and pH. Usually the sample is cooled through a sample cooler when taking samples (otherwise you risk some of the water flashing off and concentrating the TDS level in the water). You can either not bother to nutralise the sample and use a factor of 0.5 to convert to TDS (factor based approx on Sodium Chloride at 25 deg C - 1 microsiemen = 2 PPM - somebody please check I've got that the right way round, I can never remember), or neutralise as suggested by metengr and use a factor of 0.7). Might be interesting to do it both ways and see if the answers are somewhere close!
But remember, this is an approximation as different salts in the water will give different conductivities. So what you are doing is an approximation anyway.
For most accurate results, measure in microsiemens/cm as you are not introducing an arbitary figure into the equation. It is also better to measure inside the boiler using a temperature compensated system. As with all instrumentation, it is best to measure where the action is (and contiuously).

 

[jonasinc]

There is no reason to use TDS. You should just measure the conductivity. Also, by allowing the sample to cool slowly, you can introduce carbon dioxide from the air, which would increase the conductivity. You should have a proper sampling system including a primary cooler within 20 feet of the sample location and adequately sized tubing to ensure a sample flow velocity of 4 to 6 feet per second.

Also, any steam or feedwater samples that you are obtaining should be with EPRI Isokinetic Sampling Nozzles. They ensure that the sample is representative of the stream. Sample taps are not isokinetic and the small diameter holes in multiport nozzles tend to plug or erode, also resulting in non-isokinetic samples.

If you need additional information or assistance with designing the system, please let me know.