Send a piece from your own bench and we run it on the M3, then report what the machine actually did with it — including the answer you might not want.
Not your finest work. The repair you keep losing money on, or the one you turn away.
Everything below in the disclosure table. Without it, whatever we run proves nothing.
Set up under the binocular microscope in the same open work area you would be buying.
Against your current method on the same repair. Your baseline is the only fair comparison.
The useful sample is the one that represents your problem, not the one that flatters your bench.
The joint that comes back to you: a shank that keeps cracking, a clasp that will not hold, a prong tip you rebuild twice.
Offcuts of the same alloy and the same section thickness. It lets an approach be found before the actual piece is touched.
Thin, hollow, previously soldered, plated, or a mixed-metal joint. Awkward pieces settle the question; easy pieces confirm nothing.
Send an equivalent from your own stock instead of a customer's heirloom. A test piece is a piece you are prepared to lose. If the only true example of your problem is irreplaceable, write to us and send photographs before anything is shipped.
A piece arriving with no history forces guesswork, and guesswork is exactly what you sent it here to avoid.
| Tell us | Why it changes what the machine can do |
|---|---|
| Metal and alloy | Give the stamp and, separately, what you believe it actually is. The two disagree more often than anyone admits. |
| Section thickness | Name the thinnest point, not the average. The thinnest section governs the whole approach on that joint. |
| Joint form | Butt, lap, tip build-up, seam re-close, porosity fill or crack chase. Each is a different problem wearing the same word: repair. |
| Previous repairs | Solder already sitting in or near the joint behaves differently from parent metal, and it is invisible until heat finds it. |
| Plating or coating | Anything applied over the parent metal reacts in its own way around the work zone and changes the refinishing left to you. |
| Stones in place | If anything is set, say what and where. Stone-set work is assessed before the machine is switched on — see below. |
| Finish you need | High polish, matte, textured or plated. It decides how much finishing you can absorb once the joint is made. |
| Replaceable or not | We handle a replaceable test coupon and an irreplaceable piece with entirely different caution. |
Which of these matters most depends on the operation itself — see jewelry repair applications for the operations, and gold, silver and platinum welding for how the metals differ.
There is no blanket answer here, and anyone who gives you one has skipped the assessment.
A parameter table describes an alloy. Your bench holds a specific object with a history, and the history is what decides the result.
Two pieces can carry the same karat mark and behave nothing alike. One was cast last month; the other has been sized twice, plated, polished thin at the back and carries solder no one recorded. A table addresses the alloy. It has never met the object in your hand.
Copy a stranger's parameters from a forum and your first live job is the trial. That job is on your bench, in your name, and it belongs to a customer standing at your counter.
The sample is set up in the open work area under the binocular microscope, on the same M3 configuration a shop like yours would order. We work from what you disclosed — alloy, thinnest section, joint form, prior repairs, stones, required finish. Where you sent scrap in the same stock, the approach is found on the scrap first. Everything tried is recorded, including what was abandoned.
We publish no test statistics, because a statistic from our bench proves nothing about yours. Judge it yourself, at the magnification you actually work at:
This path is only worth anything because a test can fail. We will not promise you a result before we have run one, and one honest outcome is: do not do this job on the M3. Some parcels we would rather you never posted:
If the honest answer is that your category of work should stay on your current method or go out to a specialist, that is what we will write. A refused sale today costs less than a machine sitting unused on your bench. See what the M3 actually is
Three things, written plainly, with no claim that your bench will reproduce them on the first attempt.
The approach used on your joint, what was adjusted along the way, and — the part most reports leave out — what was tried and abandoned.
Images of the joint and, deliberately, of the area beside it, so you can see the finishing left behind rather than a flattering close-up.
One of three: it suits the M3 as it stands; it suits it with a different configuration or preparation; or this work does not belong on the M3.
Shipping arrangements in both directions are agreed with you in writing before anything is posted. We publish no turnaround claims here, because a schedule invented on a web page is worth nothing to you.
A sample result is not a purchase order. It is evidence for four separate decisions, and one of them is to walk away.
Describe the sample first. We reply with what to send, how to label it and how shipping is arranged — before anything leaves your bench.
Prefer email? Write to [email protected] and describe the piece before you pack it.
Not ready to post anything yet?
Ask us about your job first