Regardless of which brand or model XRF you are using, I often get calls from clients asking for help for errors, faults or bad results with their XRF. The call is usually along the lines of, “Hi Simon, my XRF is not working properly. I think there is something wrong with it.”
Being such a complicated piece of equipment with countless intricate components, complex calibrations and other variables associated with the sample itself and the user, there are just SO many things that could be going on. Whilst any good service department can help with these sorts of issues, sometimes it will require the instrument to be sent in for assessment. Let’s see if we can try to avoid this, and help you trouble-shoot your XRF analyser yourself.
There are some steps you can take yourself to rule out the majority of the errors (which, by the way, usually rest with the user or the method employed).
Sample Presentation
Let’s leave alone the whole idea of ideal geometry (placement) of the sample to the analyser. There are integrated safety features in each certified XRF to prevent a user shooting themselves with X-rays. One of them is a safety interlock that cuts off the X-ray beam if no sample is deemed to be present in front of the XRF window.
This safety feature uses a minimum count rate of X-rays back to the detector as the deciding factor as to whether the sample is present sufficiently or not. If there is not a minimum count, then the analyser kicks off the beam.
But a non-present sample may not be the reason why this safety feature is activated. It could be caused by a contamination of dust on the detector blocking the X-rays from reaching it. It could be a barrier in front of the sample preventing enough X-rays coming back.
You can check the presence of a contaminant by running the silica blank. If you get anything other than Si, you have a contaminant and you should change the window. Contamination inside means you need to send in to service.
Batteries
Li-ion batteries are notorious for depleting non-linearly. This means that when the battery meter says you have 30% left of charge, your battery may be about ready to die.
Check the battery and change out if necessary.
Ghost in the Machine
Funny how often this works – turn the instrument off and back on again and you will be surprised as to how many weird little errors disappear!
Hardware problems
Now this is the big-ticket item – any hardware errors need to be investigated by an authorised service team, so you will need to send this back in.
However, you can run the following checks to see if there is a hardware error:
- Energy Calibration – run the SS316 calibration standard using the Calibration check function in your XRF and this will tell you if the hardware is running as expected. Pass? All good. Fail? Send it in.
- Blank CRM – if you run the blank quickly in your Geochem / mining mode (10 seconds per beam is good enough), you will quickly be able to see other elements than Si if they are present. If they are there, check the blank that it is clean, check the window and change if required, or having done both of the above, it could require a clean on the inside. Service should be the only ones to open the instrument up for this.
- Known CRM – you should retain a CRM containing your elements of interest, preferably in a similar mineralogy to what you are dealing with in the field. OREAS have a good range. When you get the CRM and while your XRF is performing well, run it and note the value for your element of interest. For example, if you are interested in Cu (copper) and the CRM returns a value of 500ppm on the XRF (doesn’t matter if this is slightly out from the certified value – so long as the XRF is within ballpark and it reports consistently), then it should always give a result for Cu that agrees with 500ppm (i.e. Can be 480ppm ±25ppm).
Need more information on XRFs? Drop us a line on 0409707816 or send me an email to [email protected]
Explore XRF analysers today.