Slag Analysis

[KevinMH]

We're looking at purchasing a Spectro iQ Gas Purge XRF for slag analysis. We're mostly making low alloy carbon steels for production of iron powders. Heat analysis usually had Mn around 0.10 %, carbon around 0.40%, Ni, Cr, Cu total of 0.10% or less. Everything else at residual element levels. The heats are melted in an EAF, transiontioned to a ladle furnace for trim additions/superheat and then poured through a tundish and water atomized.

Anyone with any experience with the Spectro and it's robustness for slag analysis? The plan is to cush the slag and then press a pellet for analysis - the binder for the powder is wax.
Thanks for any feedback.

 

[steelmtllrgst]

Just purchased an ARL for slag here. Always had good results/lifetime/service with ARL products.

Have Spectro machine for OES remote analysis and the biggest turnoff is customer service, so we decided to stick with ARL when it came time for XRF slag analysis.

 

[KevinMH]

Thanks for the reply - we actually have a wide range of spectrometers here 3 different OES spectrometers: an older ARL 3460 that is a work horse, a newer ARL Quantris, and a middle aged Spectro - works well, but we've used an after market service firm for it ever since I've been involved in 2006. We also have a Phillips X'Unique XRF - pretty solid instrument, but no longer supported by Panalytical as far as ccircuit boards. We have a new Axios from Panalytical on order. The goal with the Spectro is real-time slag analysis at the generation site,otherwise I'd probably push for upgrading the Axios with a slag software package.
The non-powder side of our operation is an induction furnace/vacuum degas operation which melts stainlesses, tool steels, cobalt base alloys, nickel base alloys, monels, and the rare copper based alloy.
PErsonal preference is the Panalytical software over the ARL software.

 

[steelmtllrgst]

I see - REAL TIME, very difficult to achieve to make in-process adjustments if that is the goal.

We have high production shop and need results near immediately, have actually started using an as-cast slag sampler (small sand mold with metal chill ring). This works just OK as the segregation of slag during solidification can be extreme. Also the XRF needs new curves built in that case as no binder is used obviously. Pellet samples time from sampling to results is closer to 10 minutes while as cast is down to below 6 minutes (IF sample can be properly aquired and cooled quickly without cracking). Process still needs work!