Archived - Another addition to the lab

[Jon Rosenstiel]

Spectrace 6000 EDXRF (Energy Dispersive X-ray Fluorescence) spectrometer. Found this item on eBay (where else?) and probably paid a little too much for it, but luckily it turned out ok.

The exciter source is a rhodium anode, 50 kV, 350 μA x-ray tube and the detector is a 20 mm^2, thermoelectrically cooled Si(Li) unit. The sample chamber can be evacuated by an external vacuum pump, which makes it possible to detect elements down to sodium. (1.04 keV)

This unit was built in the early 90’s and was originally meant to be computer controlled, probably DOS or Windows 3.1. It came with a mangled ten-foot ribbon cable terminated with a 50-pin ISA connector. I did e-mail Thermo Scientific (they’ve taken over Spectrace) asking about software, but got no response. Probably just as well, had I obtained the software I would have then had to find an ISA bussed computer somewhere.

Once it was determined that getting the original software wasn’t an option I dug into (hacked may be the proper term) the units control board and rewired things for manual control. (See front panel pic).

George Schmermund walked me through process of getting the detector (which had leaked up enough that its appendage ion pump wouldn’t start) pumped back down. (See pic below).

Photos:

#1: Pumping down the detector using my fusor’s vacuum system.

#2: Liquid nitrogen cold trap.

#3: Overall setup. The unit on the bottom shelf is the spectrometers computer interface unit. It now supplies detector preamp power and detector bias, and contains the high temperature bias shutdown circuitry. The preamp’s output and pulse-reset signals have been cabled into the PGT 347 NIM spectroscopy amplifier. The spec-amps output is then fed into the Canberra Series 35+ MCA.

#4: Close up of the front panel. Originally all it contained were the “X-rays on” light and the “open” button.

A brief explanation of the controls and meters….
Chamber vacuum switch: To the left, starts the vacuum pump. (Pump can be seen on the floor in pic #3). To the right energizes the vent solenoid.

Open button: Release for the x-ray chamber’s lid. (BTW, all of the unit’s interlocks are still fully functional).

Filter switch: Controls x-ray tube filters. To the left switches in the rhodium filter, and to the right switches in the copper filter.

X-ray power switch: Energizes the HV supply. (There is a key switch on the rear panel that energizes the x-ray tube’s filament power supply).

Tube Fil current: X-ray tube anode current control… controls anode current by controlling the x-ray tube’s filament current.

Tube anode voltage: Controls output of HV supply.

Meters: Self-explanatory.

#5: Close-up of the test chamber. To the left is the x-ray tube and its two solenoid-operated filters, plus a third manually controlled filter. (Swung out of the x-ray beam in the photo). To the right is the detector and its tungsten-nickel-copper collimator.

#6: Test chamber with specimen plate in position. There should be a very thin (0.2 mil) piece of polypropylene taped over the central hole to prevent debris from falling onto the x-ray tube or detector’s beryllium windows. (Gotta’ find me a source for that super-thin polypro).

And it all works fairly well. Detector resolution (on the Mn 5.9 keV K-alpha x-ray) is ~275 eV, not as good as the originally advertised 185 eV, but not that bad either. Spectra to follow.

Jon Rosenstiel

[bwsparxz]

Nice Jon ! The chamber parts look similar to the disfuct Kevex XRF I scraped out a while back. I really need to get some N2 and give my SiLi detector a try. Converting it to peltier cooled is a thought. I'm sure the vacuum is down ( or should I say UP) in mine , but has an onboard mini ion pump. Too bad the XRF and XRD units did not have the xray tubes in them. Probibly FDA regulations on scraping units like this. May need some advice when I try mine out. Look forward to future posts on this.

[Steven Sesselmann]

John,

Nice toy, I had a chance to play with a machine like that at a recent jewellery trade fair. It was being sold as a method of doing a non destructive gold assays on jewellery alloys.

...but it could do so much more than that, the software could tell you the percentage of every element in a sample.

Have fun with it..

Steven

[DaveC]

Jon -

Very nice, indeed! Glad you were able to get the detector restored.

About a year ago, I had to find a video card for an old VLB Windows 3.1 system. That was pretty hard, but I was successful, eventually. In the process, though I found a few folks who seemed to have lots of ISA bussed hardware. Place in Texas, whose name I have somewhere... but can't recall at the moment...had a lot of hardware.

Can't help you on the software, though. And quite likely, the manual operation will be sufficient.

Be very interested to see how it performs.

Dave Cooper

[Richard Hull]

Super! What a fun find!

Our local scrap yard dude has a modern pistol that does all that.
He paid $20,000 for the pistol and a week instruction.

The source is a mutli-mCi Am241 emitter. He had to send his son to school to become the default, on-site, rad safe officer and authorized entity to safegaurd and control the source in the gun. (Fed regulation)

The gun has a large LCD screen on the back and can be forced to belch out six different screens of data. He usually keeps it strictly on a Percentage ID by element top to bottom.

He points the gun at an unknown chunk of metal in the yard (~2" away) pulls the trigger and in under 1 second he has a screen reading like.....

Copper 95.09%
Beryllium 3.10%
Tin 1.32%
Iron 0.28%,
Silicon 0.20%
trace Phosphorous, Sulfur

Cool!

Another screen will read a true spectrum with relative height spikes at mass points, etc Another will name the metal for him. Using this one function on the metal above and just pulling the trigger would return simply "Berylium-Copper 2912" another chunk of metal would report "304 Stainless Steel" or "4140 steel"

One time, he pointed the gun at a huge 1" thick sheet of filthy lead and it reported "Impure lead" flipping to percentage screen we saw

lead 98.6%
zinc 1.1%
silicon 0.15%

Scraping off the silicacious yard dust we saw a shiny surface and the lead had apparently been surface galvanized for some reason.

Super cool!

It just doesn't get anymore simple, portable or mindless than this! Just squeeze the trigger and you know to nearly .01%

You have the older larger version of a similar XRF system.

Good luck in working it. Wish I had one.

Richard Hull

[Jon Rosenstiel]

As an attempt to keep this post at least somewhat on-topic I’ve included XRF spectra comparing my fusor’s viewport glass to a chunk of lead glass that Wilfried Heil graciously sent me.

Pic #1: X-ray tube anode voltage of 18 kV, test chamber under vacuum, live time of 200 s.

Pic #2: X-ray tube anode voltage of 50 kV, test chamber under vacuum, copper filter switched in, live time of 200 s.

Jon Rosenstiel

[behrooz]

Hello Jon,

We also have got one but it has no software. Could you please upload the software you have used somewhere for me to download or use the following link for uploading to my dropbox?

https://www.dropbox.com/request/fx4td8HV23tD9a9j5ZDS

Agoo

[Jon Rosenstiel]

I didn’t get the computer interface or software or manuals, sorry.

Jon Rosenstiel