Helium analyzers

[TwoBitTxn]

How does a He analyzer work?

I understand how an Oxygen analyzer works, but He is an inert gas.

Would someone explain it please.

TwoBit

[Bottomfeeder]

TwoBitTxn once bubbled...
How does a He analyzer work?

I understand how an Oxygen analyzer works, but He is an inert gas.

Would someone explain it please.



It compares the thermal conductivity of the sample gas to that of a known reference. It uses heated resistance thermometers (this is similar to the mass flow sensors in cars). See http://www.atomox.com/he_tech.html

Helium is much more conductive than nitrogen and oxygen.

[omar]

The measurement technique is a pair of Thermal Conductivity Cells in a Wheatstone Bridge circuit.

To determine gas composition by thermal conductivity each gas must have a significantly different thermal conductivity to be able to differentiate the amounts. Because the thermal conductivities of Nitrogen and Oxygen (N2 and O2) are very similar they can be considered to behave as a single gas. Hence, a mix of oxygen, nitrogen and helium can be considered a binary mix for thermal conductivity measurements. The thermal conductivities of Air, N2 and O2 are: 26.2, 26.0, 26.3, milliwatts per meter Kelvin, and the thermal conductivity of Helium is 156.7 mW/mK. (Values for 300 degK/27 degC)

The technique is not suitable for measuring extremely low levels of gas (say, <1% by volume resolution), and it has difficulties when the gas to be measured has a thermal conductivity similar to the background gas. Thus measurement of oxygen in air is not feasible because oxygen has essentially the same thermal conductivity as air. However, for certain gas mixtures it is an extraordinarily useful method of measurement.

In addition, the conductivity of a gas mixture is always less than a weighted average based on the number of moles (or molecules) of each component in the mixture. This appears to be related to the different effect that molecular weight has on thermal conductivity and viscosity. Light gases are usually better conductors than are heavy gases, whereas heavy gases are often (but not always) more viscous than are light gases. There also seems to be some correlation between molar heat capacity and thermal conductivity.

Because of this theoretical models turn out to be poor predictors of the behavior or response of thermal conductivity analyzer to a gas mix. This is not unexpected. The science regarding thermal conductivity of gases is still some times referred to as black magic because there are inadequate explanations for the actual behavior of gases.

omar

[TwoBitTxn]

I understand how it works now.

Omar,

If I understand you correctly, at low levels thermal conductivity meters can have accuracy problems.

At helium levels trimix divers are using are thermal conductivity meters accurate or is there another way to meaure the He in a cylinder.

TwoBit

[omar]

The low levels I was referring to are less than 1% or ppm (parts per million) range. A trimix of only 1% helium would in effect be the same as air.

I know that the ATOMOX HeLIUM ANALYZERS http://www.atomox.com
are accurate +/- 1% (actually better).

There are other techniques to determine helium contents such as the speed of sound or the optical wavelength of the gas fractions.

omar

[oxyhacker]

An interesting distinction is that, while an O2 analyzer is actually reacting the amount of O2 in the sample, a He analyzer is just surmising what the He content is based on the assumption that the other gases are nitrogen and 02. That is to say, if you inadvertently feed some other mixture of gases to it, an He analyzer will dutifully attempt to report its %He even if there is no He in it at all. This shouldn't really be a problem in real life, unless you are keeping bottles of hydrogen or maybe neon in the mix room.

BTW, the He analzyer game is heating up.
http://www.oxycheq.com
has a nifty new pocket sized He analyzer. These things are quite simple mechanically once you get the electronics sorted out; it will be interesting to see what effect competition has on the price.

[omar]

LOL Vance…. (heating up ;-)

But, it seems with the introduction of the oxycheq the price has gone up. The oxycheq MSRP is more than the Atomox distributed by Dive Rite and you have to buy the smart charger on top of that.

As far as operation of the oxycheq you need to screw around opening and closing the flow cap to capture a gas sample every time you want a reading, it takes a minute to get each reading, you need to turn it on and off between readings to save the battery, you can not use the smart charger when the unit is on, and no mention is made if running a high oxygen mix through the sensor/flow cap is safe. So as far as using it for continuous blending it is very limited.

When I continuous blend it is for a number of hours. A limited battery life and not being able to hook it up to the main power supply is a detriment. During start up I need continuous, instantaneous readings of the mix so I can get it dialed in without wasting a lot of gas and the last thing I need is to have to worry about capturing a gas sample for a reading. Atomox recommends the sample flow rate be 2L/min but has indicated that the accuracy of the unit will still be within +/- 1% for flow rates of 1 to 12 L/min. So all I do is turn on my unit, plug in the gas and that's it.

The oxycheq is smaller and more portable which would be a benefit if you were mixing on a boat.

omar

[Uncle Pug]

omar once bubbled...
The oxycheq is smaller and more portable which would be a benefit if you were mixing on a boat.

Well I don't think the Atomox is that big a package to take on the boat at all... I mean if your boat is big enough to be tech diving off of then that little pelican box is nothing... alas, I do not have a battery pack for mine... just a lowly adapter that plugs into the wall.

But really... who needs the He analyzer for boat mixes anyway.... just use PP blending and analyze your O2.

The Atomox is the ticket for Continuous Blending.

[AquaTec]

I have the atomox and the OMS O2 analyzer, they both fit ncely into the same pelican cased the He analyzer came in. and btw if you take the cover off the He analyzer you can put batteries in ther. I ahven't done that yet but i understand the batteries willl last quite a while

i find the He analyzer most usefull for air top ups for your second dive on the same tanks

[padiscubapro]

omar once bubbled...
LOL Vance…. (heating up ;-)

But, it seems with the introduction of the oxycheq the price has gone up. The oxycheq MSRP is more than the Atomox distributed by Dive Rite and you have to buy the smart charger on top of that.

As far as operation of the oxycheq you need to screw around opening and closing the flow cap to capture a gas sample every time you want a reading, it takes a minute to get each reading, you need to turn it on and off between readings to save the battery, you can not use the smart charger when the unit is on, and no mention is made if running a high oxygen mix through the sensor/flow cap is safe. So as far as using it for continuous blending it is very limited.

When I continuous blend it is for a number of hours. A limited battery life and not being able to hook it up to the main power supply is a detriment. During start up I need continuous, instantaneous readings of the mix so I can get it dialed in without wasting a lot of gas and the last thing I need is to have to worry about capturing a gas sample for a reading. Atomox recommends the sample flow rate be 2L/min but has indicated that the accuracy of the unit will still be within +/- 1% for flow rates of 1 to 12 L/min. So all I do is turn on my unit, plug in the gas and that's it.

The oxycheq is smaller and more portable which would be a benefit if you were mixing on a boat.

omar

Nope..

If my partner sees this he'll probably be pissed (he has a good friend that works for DR) but we now have any Oxycheq one in the store.

I have used both.. I now have the Oxycheq analyzer (actually I was the First customer to get one)..
The Oxycheck analyzer does need to be charged. It lasts a few hours on the charge but the charger comes with it..
the test time is very fast..all you have to do is flow gas through for a few seconds(at low flow rates 5 seconds is more than enough).. if you have a high flow rate its basically instantaneous(the flushing).., push in the sampler and wait.. on average you will get a stable reading in under 15 seconds.. I find mixes up to 50% average well under 10 seconds(its within 1 % in under 3 seconds).. the atomox takes much longer and uses much more gas for most uses.. plus you have to let the unit go through a long heatup (4 minutes if I remember correctly) before you can take a reading.. IF you like a unit like that oxycheq will also be selling a unit with built in O2 made for him by teledyne and will have all the same drawbacks as the atomox..

I have tested it(oxycheq) on mixes from 0% helium (which the oxycheck reported 0.1) to (100% helium which it reported as 99.8) the atomox can't test for less than 4% or over 90% (actually I have never seen it read 90% on mixes that should be 90%) I did a 90-10 heliox were my o2 analyzer said .10 but the atomox said 89% (still witihin it error range) but clearly not as accurate as Oxycheq's.
If you aren't traveling with a He analyzer weight wount be an issue.. I travel with lots of gear and every pound I save (it would be several) is welcomed.

Even if you are continuous blending trimix there is no reason to have to constantly monitor He content.. O2 can be constantly monitored (as a precaution) if it changes you know the He is changing.. with good flow regulators there should be no drifting once you get your mix set.. the better regs are compensated for drop in tank pressure.


I will do some detailed timed testing that will probably be shown in AdvancedDiver magazine.

[Uncle Pug]

padiscubapro once bubbled...
Even if you are continuous blending trimix there is no reason to have to constantly monitor He content..
...just to get the initial settings because they change with ambient temp. and then I check every so often... because as the compressor runs the temp. changes cause a slight drift in the mix. Sometimes I adjust for it and sometimes I don't.... but I like to know.

[padiscubapro]

Uncle Pug once bubbled...

...just to get the initial settings because they change with ambient temp. and then I check every so often... because as the compressor runs the temp. changes cause a slight drift in the mix. Sometimes I adjust for it and sometimes I don't.... but I like to know.

I agree and if the compressor is not in a confined area temp should be fairly consitent. If you see the o2 changing, then a check would be necessary.. but monitoring both is not necessary.. and if they are set up in a flow through design both can be checked whenever you want to do a check..

The oxycheck analyzer is set up so you can flow through the gas all the time.. if you decide to take a measurement turn the unit on, push in the sampler (since its already flushed) wait a few seconds and you got a reading.. pop out the sensor and let gas flow again... seems like a good arangement to me..

And this arrangement doesn't cause any problems setting it up pass the gas through and see you o2 settings catch a sample pop out and make another adjustment... and repeat.. it shouldn't add to your time and the gas sampling for helium is non destructive, so you put it first in line, no need to run multiple sampling ports, 1 will due, if you put a check valve on the sample port the sensors will be isolated.

[omar]

I have tested it(oxycheq) on mixes from 0% helium (which the oxycheck reported 0.1) to (100% helium which it reported as 99.8) the atomox can't test for less than 4% or over 90% (actually I have never seen it read 90% on mixes that should be 90%) I did a 90-10 heliox were my o2 analyzer said .10 but the atomox said 89% (still witihin it error range) but clearly not as accurate as Oxycheq's.Interesting… How do you know that the oxygen was correct? For this comparison you could just as easily say that the atomox was right and the oxygen was off by 1%. If you mixed by ideal gas amounts then the helium contents would be less than what is expected/predicted.

Run a 50% EANx (no helium) through the oxycheq (still don't know if this is safe). What you will get is about 1.5% readout. Using this value then would indicate that the oxycheq is not as accurate. My point here is that you can not use this values as an indication of the accuracy. If you are going to test them you need to use calibration gases prepared by a lab and do them to manufacturers recommendations. If not the test will be flawed.


Even if you are continuous blending trimix there is no reason to have to constantly monitor He content.. O2 can be constantly monitored (as a precaution) if it changes you know the He is changing.. with good flow regulators there should be no drifting once you get your mix set.. the better regs are compensated for drop in tank pressure.Hmmm… if you take more than 10-12 readings of the mix during the process it will take a greater amount of time to get them with the oxycheq (this is with your 15 sec reading). And you still have to open and close the flow cap on the oxycheq which in my opinion is less than ideal. Once the atomox is warmed up you get an instantaneous & continuous reading.

You will get drift in the mix for a number of reasons. The input of the compressor will be less when it is pumping at higher pressures, temperature variations, and the output of gas cylinders should be controlled by flow meters (if you are using just the regulator output it is very had to fine tune the mix) and even good ones will drift slightly. As I said it is during the startup that the most adjustment is needed to get the mix dialed in, after that it just needs minor tweaking.

Also, if you are taking the precaution to monitor the oxygen all the time then why not for helium as well? The atomox will let you do that with no hassle.


I will do some detailed timed testing that will probably be shown in AdvancedDiver magazine.As I said before, if you are going to test them you need to use calibration gases prepared by a lab and do them to manufacturers recommendations. If not the test will be flawed.

Also, Teledyne has been making thermal conductivity analyzers for years and has a wealth of experience. It is not surprising that the dual unit they will make will be functionally similar to the atomox. Teledyne's current units require an 8 hour+ warm-up/equilibration period for the unit and the gas(es). (This is with the gas flowing!)

omar

[padiscubapro]

omar once bubbled...
Interesting… How do you know that the oxygen was correct? For this comparison you could just as easily say that the atomox was right and the oxygen was off by 1%. If you mixed by ideal gas amounts then the helium contents would be less than what is expected/predicted.
omar

Omar, I mix using real gas laws and so far everything that I Have tried to mix and validate using the He analyzer has been almost dead on(usually within .2).. ON my banked mix, I mixed it using real gas and the volume is such that its slightly more than 1cft/psi so its quite easy to mix accurately.. On the heliox mix I verified the oxygen content with 2 different meters that both agreed with each other.

I have the batch testing results for the 100% helium (it was virtually 100% with a few PPB of argon)I used as a referance so That test is valid.. Air has so little trace He that that test should also be valid... I will try it on a 50% nitrox mix and see what I get.. I am also going to check it against some known premixed gasses from our supplier and see how it does with variuos other gases present..

The atomox may or may not be better for permanent use since you can monitor it continuously but its too big and heavy to carry with you everywhere. Their new unit looks like they are trying to make it a better option, but its still heaver and you need a palm pilot...
Personally whether its off by 1% or 5% its not going to be significant in your decompression, since people were diving untested mix for years and we didn't see lots of people getting bent.

the analyzer is much more important when you are trying to use the leftover gas and not start from scratch.

if you are using a back pressure regulator on your compressor (which is highly recommended to maintain dry gas) your compressor would be presented with close to the same pressure at all times..

[Patrick]

Omar,

The suggested retail is the same as it was. People that attended DEMA were able to get them at wholesale prices. The Expediton He analyzer comes with a smart charger, it is not extra. We were told by many people we were the busiest booth at DEMA ... other than maybe the magazine publisher that was handing out free beer.

The OxyCheq Expedition He Analyzer is very simple to use ... flow the gas, push the sensor cap in, turn it on, and wait a minute for a very accurate reading. The unit is temperature compensated, has no warm up time, orientation is not a concern and various orientations do not effect accuracy. The Expedition He Analyzer can be used with the smart charger (that is included in the price) plugged into the unit if desired.

Prior to design, we spoke to commercial diving operations and they check He every 20 minutes. So, every twenty minutes, you push in the sensor cap for a reading. We claim a reading within 1 minute (most people will say it is much less). Lower He percentages is much less time. The Expediton He Analyzer can also measure He from 0-100% and the accuracy is not effected by orientation or gas flow. If hooked up to a contiuous flow system, you CAN keep the charger plugged into the unit and leave it on.

Running 50% oxygen through the Expedition He analyzer is not a concern and we do not indicate accuracy, we state it.

I did not design the Expedition He Analyzer, I had people that have been in the business for many years design it. Anyone that has been involved in this type of gas analyzer will tell you that units with continuous flow rates will change accuracy if you are outside the target flow rate. Often orientation and temperature changes will effect accuracy as well. The Expedition He Analyzer captures a sample of gas and this is why orientation, gas flow and temperature is not a concern for the unit. It is also why we have no warm up time. We think these obvious benefits out weigh pusing the sensor cap in when you want an accurate reading.

If anaylzing single tanks, I believe the benefits are even more pronouced.

My best advice would be to ask people that have used both units tell you what unit they prefer. I only say that because I think it will result in more sales for me. :)

If anyone on the list is interested in an OxyCheq He Analyzer, let me know and I will put you in contact with a dealer that can help you and even ask them to give you a special price.

Thanks for giving me the opprtunity to clear up some misconceptions or information that is not listed on the OxyCheq site.

Best regards,

Patrick
OxyCheq
http://oxycheq.com

[padiscubapro]

Patrick,
I guess my battery was too dead when I first got the unit and was taking too much of the available current.. I tried working it with it plugged in... I'll try it again for continuous monitoring.. So far its working like a charm.

Thanks for the clarification.. I've been just using it battery operated and recharging when I got battery low indication..

That call I suggested that you make... the person will not be back in the office until 11/21.. talk to him and let me know what you are going to do..

Joe

[padiscubapro]

OMAR,
contrary to whats posetd on a certain website the Atomox unit is greatly affected by flow rate.. I had someone that I know try doing some measurements with flow rates from about 1lpm to about 15lpm (I say about because you CAN'T be sure because you don't know the mix to start with, flow meters calibrated for air will be wrong when showing flow with anything but air, the greater the percentage of He the more the flow meter will be off by.. I have a table somewhere which shows you how to figure out the actual flow rate)
The atomox gave readings that varied over the entire range... and this makes sense because when you design an analyzer what you do is use a calibrated flow rate and compare your heat curve base on one of several known samples over a range, its only valid for THAT flow rate, alter the flow rate and the numbers are different.. if you allow VARIOUS flow rates you would have to be able to tell the meter what the flow rate is.. The higher the actual He content was the more the sample analyses varied.

The only other way is to use a calibration gas to calibrate the meter based on the flowrate you are going to use.. so that the analyzer could figure out where this flow rate fits into its heating curve.

Designing a He analyzer is actually quite simple, its bassically a wheatstone bridge with a thermally active device that will cause a varyance in a voltage. then you just need calibrated samples to build your curve..