Generating plateau curves

[cincinnatinuke]

Hi folks.  I have gotta a dilema here at work.  I am being asked to bein generating plateau curves for an Eberline mini scaler models MS-2 and MS-3.  First I only have a manual for the MS-2 so if you could help locate one for the MS-3, many thanks.  I have looked online and visited the website for Thermo Electron(Eberline) and to no avail.

Secondly, in the manual for the MS-2 it appears that no plateau has to be done for a GM probe but only for a proprtional and scintillator probes.  This particular scaler has many different probes that are compatible.  So, a theory question....do GM tubes not need Plateau curves?  We did them in the Navy and had GM tubes.  Any help would greatly be appreciated.

[astronuke]

You would normally plot a plateau curve the first time a GM detector is used to determine the optimal operating voltage.  After the instrument has been recalibrated, you would normally just use that optimal voltage unless there was a reason to suspect that a new platueau was needed (i.e. significant adjustments to the MS-2 during calibration).

[cincinnatinuke]

That is part of the problem.  There is no record for an initial curve ever being done and now Ohio Dept. of Health ( We are an agreement state here in Ohio) wants them done.  To do them is no big issue, but the lack of an initial one prompted me to the tech manual for the MS-2.  Upon the initial setup it states nothing for curves, only to select the desired voltage on one axis of a given graph and then be given the dial setting for that particular voltage.
I have no real reason to suspect a problem with the tube.  In fact we have sent it off to be calibrated in the past, and me being the F.N.G. said we can do that and save us the money we'd spend.  A few months later the task is mine since cals are due this month. 
Another issue is the detector and the tube are 10-15 years old.  Seems like back in the day they were replaced more often than that.
Also any word on how to get a tech manual for an MS-3???
C Nuke

[raymcginnis]

I have been doing instrumentation for years now.  Plateaus are a theoretical knowledge from the old days.  The reality is that the plateaus are something that has been proven time and time again by the manufacturers.

If I had to guess, I would almost guarantee that the GM tube that you are using likes to operate at 900 volts.

You don't have to spend the time proving what has been proved over and over again.  Set it for 900 volts and then calculate your efficiency. 

Plateaus makes a nice test question, but you don't have to do them on GM tubes anymore.  It has been done for years now.  900 volts for almost every GM tube.  I hope this helps.

[alphadude]

and ramping up the tube voltage to get the curve doesnt do a lot of good for the tube. when you got to cd you are "cooking" the tube   

[cincinnatinuke]

Thanks for the info guys.  I still need to generate one since the original was either not done or just not documented.  I am firguring it was just plain lost. 
If anyone out there with an MS-3 knows how to calibrate the count times ( i.e. to ensure a one minute count is one minute and so forth) I'd appreciate any info.  I did check the search posted on google and did my own again and nothing.  I am still waiting for a call from Thermo or even an email regarding this so maybe it will work out as far as getting a replacement manual.
Thanks again and this site truly is awesome.

[raymcginnis]

A simple stop watch should do.  Start the counter and the stop watch at the same time.  When the counter light goes out, stop the stop watch and documnet the time.  This is not perfect but it gets you within the range every time, if the counter timer is accurate, which is what you are trying to prove.  So you may get 59 seconds or 1 minute and 1 second for a 1 minute count.  That is passable.

Sounds like you are reinventing the wheel.

[alphadude]

you need some sort of NIST traceable device to measure the circuit output for the timer. generally those timers are pretty sturdy but it you are doing this as part of the calibration you need to scope it against a NIST device or have a certified technician service the counter. If you want to do this for your own peace of mind-like roy said get a watch.

[tabai]

At our Facility we do Plateau curve for Scintillator Probes only. NOn Scintillator probes areleft at 900 Volts.

[cincinnatinuke]

On our scintillator probes we perform a procedure similar to plateau curves, with a manipulation of sensor voltage vs. counts, but we dont expect a plateau in its shape.  For scintillator probes we are looking for a max output within the operating range.  It doesnt vary much throughout the range, but the hope is to hit a "sweetspot" of sorts.

On a simialr note, I notice that when initially powering up the unit and bringing voltage up the first count I run I see a lot of discharge.  You can hear the tube popping and counts jumping in intervals of 3,4, or 5 instead of by ones.  The slower the voltage is brought up the better, but not by much.  I consulted the manual and no instruction is given on how quickly to bring voltage up.  I never saw this in my ELT days, but of course I am dealing with background near 10-15 CPM.

Well to the point I was trying to make before I started rambling on.  I was asked to explain variances in background from one day to the next or even from one count to the next.  So on I ramble about statistical variance and the like, showing that flucuations were within x amont of standard deviations.  But then I started to think this evening about other factors like time of day, weather, humidity.........there were others that were more of a factor and maybe this whole Katrina/Rita thing got me.  I wonder how much the "elements" around us affect sample counting.  Heck I remember seeing differences in backround on a Sub in port vs. underway.  So if to add nothing more than to further this topic in a new direction I offer my random thoughts on this.

[alphadude]

We noticed a few percent increase in eff. when we move to higher altitude. Rocky Flats gave us about 3% more eff. due to air density than at sea level. Humidity would also factor in.

[LaFeet]

I dont recall doing plateau curves for GM probes while in the Navy... when did that start?

[Holland32Inst]

Depending on the GM tube manufacturer, the operating voltage may already be posted for you.

For example, for LND tubes; http://www.lndinc.com/gm/gmpan.htm (voltage dependent on tube size/dimensions).

There are other tube mfrs. TGM, etc...

Or, on rare occasions, I have even seen the gentle, simple, "education" of the regulating body member concerning the gas amplification curve. Much tact and diplomacy required.

In my Navy days, the Navy's pancake GM tube calibrations were cookbook. The voltages were predetermined, the meter assy was electronically pulsed and the GM tube response was verified that it fell within a predetermined range (pre-determined response range based on desired efficiency). Part of the reason being, that there is little efficiency change for the operating region of the gas amplification curve for GM tubes. Pulse height is not a major factor. (One large pulse from the tube per event vs gas proportional tubes where pulse discrimination is required based on pulse height, usually to discriminate for alpha vs beta, i.e. Eberline PCM) GM region vs the gas proportional region.

Other GM tubes for dose rate, (remember the old AN/PDR-27s, and AN/PDR-43s?) I'm dating myself, are also set to a pre-set voltage and tweaked to the desired response in a rad field.  Again, no plateau required.

Scintillation/photomultiplier tubes are a different animal altogether.

[LaFeet]

So how much af a difference does a 10% change in voltage alter the output signal.  Is this not why we use this section of the Gas Amplification curve for GM tubes?   

[cincinnatinuke]

I didnt mean to imply that plateau curves were done on survey meters with GM tubes nor do I remember doing response curves that I see now in the literature that accompanies meters.  The plateau curves I did in the navy were for the counting "pig"(for primary samples and the like) we used when a tube was changed out.  I dont remember generating a curve but only verifying that response vs voltage was within calculated tolerances. 
I do remember in prototype that we were given some KAPL or Bettis handout that talked to some detail about plateau curves and associated equations.  i wish it would have been NOFORN so I could have kept it.  I did however get to keep some engineering principles books that were given.  It serves as a great reference tool or just plain sleep medication.
The gentleman from the regulating body had made the recommendation to verify functionality of our counting equipment for contamination surveys.  The biggest problem arose in the lack of an established program being written up, with references to established procedures and practices, and hence I dove into re inventing the wheel as one of the earlier posters had noted.  This was an observation from before my time at my current employer, but nonetheless it was determined to write up procedures for everything we could possibly think of.  Many headaches later, I wish I could have just gently educated the regulator, or at least been there to ask what exactly would suffice.  Since I was not, I am now in the midst of quite a large project with no end in sight since more gets added to the pile.  But it is nice to be able to take a breather and view posts here on nukeworker and see that many are out there with just as many questions as I or even better is to see responses to my post.  Once again thanks guys!!

[LaFeet]

The plateau curves I did in the navy were for the counting "pig"(for primary samples and the like) we used when a tube was changed out.  I dont remember generating a curve but only verifying that response vs voltage was within calculated tolerances. 

I recall doing that for the SMAGs when they needed it.   I allways enjoyed working with the mechanically and electrically challenged 

[cincinnatinuke]

Speaking of electrically challenged I remember being in nucleonics in prototype and watching someone accept the challenge of 1000 VDC to the CDA chamber.  It went more or less like "DC Cant hurt you only AC can" to "You wanna bet?" to....well you get it.

[LaFeet]

Speaking of electrically challenged I remember being in nucleonics in prototype and watching someone accept the challenge of 1000 VDC to the CDA chamber.  It went more or less like "DC Cant hurt you only AC can" to "You wanna bet?" to....well you get it.

Oh my Gawd... ROTFLMAO

I remember having to reinstall the door to nucleonics for this very reason... I also got pelted by the LELT