Elvatech Ltd.

24/09/2026

Three days in Kielce are done.

The whole time we asked people to pull out their own metal. A bolt from a pocket, an offcut, a part someone had doubts about. Every analyzer looks good on a perfect sample. The truth starts on a rusty bolt.

Thanks to everyone who came by and argued with the number on the screen.

21/09/2026

Tomorrow.
METAL 2026 opens at Targi Kielce Targi Kielce S.A. and runs through Thursday. We are at stand 6-B04 for all three days.
If you are coming: bring a sample. It remains the most useful thing you can do with five minutes at a trade fair.
If you are not coming but have a question — write to us anyway. Being at a fair does not stop email working.
See you in Kielce. 🇵🇱
📍 Targi Kielce · 22–24 September · Stand 6-B04 🔗 https://f.mtr.cool/ia8009336e

20/09/2026

Two days until METAL 2026 opens in Kielce. Targi Kielce S.A.
A quick note for anyone still deciding whether to bring something to stand 6-B04: it does not need to be interesting. Ordinary material is often the most useful thing to test, because ordinary material is what you buy every week without checking.
A routine delivery. A grade you have never questioned.
Returns that have been sitting in the yard long enough that nobody remembers where they came from.
Those are the samples where a ten-second measurement occasionally changes somebody's Tuesday.
📍 Targi Kielce · 22–24 September · Stand 6-B04 🔗 https://f.mtr.cool/v8yl7n8agt

19/09/2026

Packing an analyzer for a trade fair is a strangely careful process.
The instruments themselves are built for foundry floors and scrap yards — sealed, shock-resistant, designed to survive a working day in a place with no clean surfaces. And yet somebody always wraps them like glassware before a drive across Europe.
Reference samples, calibration standards, spare windows, batteries, the small box of things nobody needs until the one time somebody does. Three days in a hall means packing for every question that might walk up to the stand.
Kielce Targi Kielce S.A. on Tuesday. Stand 6-B04.

18/09/2026

Four days out. Here is what will be on the stand.
ProSpector 3 — handheld, SDD detector. Alloy grade in 2–3 seconds, full quantitative chemistry in 5–10. Goes to the material: the yard, the charge bay, the floor. 1.5 kg, eight-hour battery, sealed against dust.
ElvaX benchtop line — fixed chamber geometry, which removes the operator from the measurement entirely. Same result from every person on every shift. For a QC room where two technicians must report the same number, that is the whole argument.
Most foundries that run both use them differently: the handheld screens incoming material, the benchtop documents what leaves.
We will also answer the question nobody asks at a stand but everybody wonders about — where XRF stops working. Carbon, for one. It cannot see it, and no configuration changes that.
Come and find the limits yourself.
Targi Kielce S.A.
📍 Targi Kielce · 22–24 September · Stand 6-B04

18/09/2026

We publish a lot of material explaining where our analyzers are the right tool. This is the other kind.

If you do federally regulated lead-based paint inspection in the United States, you cannot use a ProSpector. Not because it measures lead badly — it measures lead well. Because of a document.

An XRF analyzer used for regulated lead paint work must hold a Performance Characteristic Sheet issued by HUD for that specific make and model. HUD's wording leaves no room:

"If an XRF analyzer does not have a PCS, or if it is not used, or if the data are not analyzed, in accordance with its PCS, the actions undertaken with it are neither a lead-based paint inspection nor paint testing."

Two instruments can produce identical numbers on the same wall, and only one of them produces a legally valid inspection. Elvatech analyzers do not currently hold a PCS. If that is your work, buy from the HUD list — the Niton XLp and XL-309 series, the Viken Pb200i, the LPA-1, and the others published by HUD and the National Lead Information Center.

We would rather lose the sale than have someone discover this when their report is challenged.

Our Head of Laboratory wrote the full guide this month, and it covers what actually applies in 2026:

→ The threshold is 1.0 mg/cm² or 0.5% by weight, under 40 CFR 745.103

→ Maximum TSCA penalties are $49,772 per violation per day — and unchanged from 2025, because the inflation adjustment never happened after the autumn government lapse. The $37,500 figure still circulating online is several cycles out of date.

→ An inconclusive result is a defined outcome, not an instrument failure. It goes to a laboratory. It does not get resolved by taking readings until a convenient one appears.

→ The end-of-day calibration check is the one people skip and the one that invalidates the day's work.

And the part where our instruments do belong: lead in soil, in consumer products and toys against CPSIA limits, in scrap and solder, in RoHS screening, in service-line inventory work for utilities. None of that is regulated paint inspection, none of it requires a PCS, and all of it is a larger field than the part we are ruling ourselves out of.

Read the guide → https://f.mtr.cool/hhhayz4suj

XRF lead testing under EPA RRP: the 1.0 mg/cm² threshold, the HUD Performance Characteristic Sheet rule most buyers miss, penalties and field protocol.

17/09/2026

Trade fair stands mostly contain things you are not allowed to touch.
Ours will not.
At METAL 2026 Targi Kielce S.A. goes in your hand, not behind glass. You pull the trigger, you read the screen. If the number surprises you, we will run it again and talk about why.
Things worth bringing:
→ Charge material you buy regularly and have never independently verified → An offcut from a batch that behaved oddly → Two pieces that should be the same grade and are not performing the same → A casting that failed and left no obvious explanation → Anything where the certificate and your instincts disagree
No appointment, no form, no queue for a sales conversation first.
📍 Targi Kielce · 22–24 September · Stand 6-B04

17/09/2026

A foundry gets one chance to be right about the charge.
Once the melt is poured, the composition is whatever it is. Adjusting after the fact means remelting, and remelting means energy, time, and a schedule that no longer works.
Which is why incoming material verification in a foundry is not paperwork. It is the last point at which a mistake is cheap.
Scrap arriving as one grade and turning out to be another. Returns mixed across alloy families in the yard. A supplier substitution nobody flagged. Each of these produces a melt that meets no specification, discovered when the casting is already cooling.
Portable XRF moves that check to the point of delivery. Seconds per piece, no sample preparation, no waiting.
We will be demonstrating exactly this at METAL 2026 Targi Kielce S.A. next week — with your material, not ours.
📍 Targi Kielce· 22–24 September · Stand 6-B04

10/09/2026

Sending samples to a laboratory looks cheaper than buying an analyzer. One ICP sample is about $100. An XRF machine is $30,000.

At 200 samples a month, the lab invoice passes the price of the instrument in under two months

That arithmetic is the easy part. The harder part is what does not appear on either invoice — the three days a decision waits, the material sitting in quarantine, the rig standing still while somebody in another city runs a digestion.

Our Head of Laboratory published a comparison this month that works through it properly: five-year costs at three real volumes, break-even calculated rather than asserted, and the cases where the honest answer is that you should keep using the lab.

Three of those cases, stated plainly:

→ Trace levels below XRF's reach. XRF sits around 10 ppm for most metals. ICP-OES reaches about 1 ppm. Only ICP-MS gets to parts per billion — a distinction that gets blurred in vendor material constantly.

→ Light elements. Everything below sodium is invisible to XRF. Lithium in battery materials, carbon in steel certification, boron in master alloys — all need another method

→ Accredited certificates. ISO/IEC 17025 accredits a laboratory, not a method and not an instrument. There is no such thing as an ISO 17025 certified XRF machine, whoever built it.

And one scenario in the article where the numbers say not to buy: a shop running 30 gold tests a month has a payback period of 163 months on test cost alone. We put that in because it is true.
Read it → https://elvatech.com/xrf-machine-vs-lab-analysis/

XRF machine or lab analysis? Real numbers: 5 seconds vs 5 days, $1 vs $100 per test, break-even under 3 months. See the full 5-year cost breakdown.

04/09/2026

Not every spot on a piece of scrap tells the whole story.

Scrap isn't a laboratory sample.

It can be thin, curved, dirty, welded, coated or mixed with other materials.

That's why measurement technique matters.

A good rule for operators:

✔ Choose a representative surface.
✔ Avoid edges, welds and obvious contamination.
✔ If the result looks strange, move to another spot and test again.

The analyzer can give you a very precise number.

But the real question is:

Did you measure the right place?

Good XRF practice is not about blindly trusting the screen.

It's about combining the measurement with operator judgment.

every spot on a piece of scrap tells the whole story.