Free Weld Sample Evaluation: One Real Part, One of Four Answers Back
Which of the four you get depends on which part you put in the box.
One of those four is that we turn your part down. That one is written out below, including what you are holding when it happens.
Start with the part on your benchFour Answers, and There Is No Fifth
Whatever your part turns out to be, it lands on one of these.
- Yes. — Proven on the part you sent, at settings someone else can repeat.
- Yes, but not there. — The there is a point in your own sequence of operations.
- Yes, and we'd still say don't. — It welds here and won't hold across a production week.
- No. — The limit gets named, and it isn't one a bigger machine moves.
Four. Your part lands on one of them.
Answer One: Yes — and What the Sheet Is Worth Once It's Back on Your Bench
The welded piece is the part you show people. The sheet that came with it is the part you use.
You already know the weld held, because you're holding it. What the sheet adds is the half that leaves your desk: an operator who wasn't there can set the machine to it, and someone who will never pick the part up can see that the result had a cause rather than a good afternoon behind it.
Without the sheet you own a sample. With it you own a step in your process.
Take it back to the joint it was written on, not to the family of joints that resemble that one. Same alloy, same section, same neighbour still attached — those are the conditions the numbers were established under, and they travel with the numbers.
Treat the sheet as the starting line: the first thing you'll adjust is usually the fixture, not a setting.
Send a second part when something upstream of the joint moves — a new supplier for the same alloy, a section that got thinner, a coating or a stone introduced after the first sample was welded. That's a new question rather than the same one asked twice, and it gets answered the same way.
Answer Two: It Welds — Just Not Where You're Making the Weld Now
The most common correct answer on this page arrives sounding like bad news about your process.
You sent a part to settle a question about machines, and the answer that lands is about a decision you made months ago: the point in your sequence where this weld happens. The joint is available. It just isn't available there.
That answer only exists because the weld was made on your actual part with the neighbour still attached to it — the same part described in an email would have got a yes, and the yes would have been useless to you.
It isn't a softer no, and it isn't a hedge. What it asks of you doesn't come out of a budget — it comes out of the schedule, and that's the part that stings.
A step that currently happens where it's convenient has to start happening where it isn't; the week it disturbs is a week that was already planned around; the people it inconveniences are your own. There is no part of it you can hand to somebody outside the building.
What it costs to act on is an afternoon with whoever owns the routing sheet, and you'll be walking into that conversation holding the sample. The objection you'll meet is that the new order breaks a habit; the answer to it is a piece of metal somebody can pick up and turn over.
If the sequence genuinely can't move — the part is already assembled, already sold in that state, already in a batch that's half-built — say so when you send it.
The evaluation then answers a narrower question: whether the joint can be reached and held at all in the state you're stuck with. That's a different test, and it gets run as one rather than quietly folded into the first.
Answers Three and Four: What You're Left Holding When the Useful Thing to Say Is Don't
Two ways this ends without a machine being recommended — and what each of them leaves you with.
Answer three · It welds, and we'd still say don't
You can have this joint welded — that's the uncomfortable half of answer three.
There is a version of it that works on a good day, with one operator who has done it four hundred times and nothing else going on at the bench. We can produce that weld and photograph it, and sending you that photograph is the easiest dishonest thing available to us here.
So what you're told instead is how narrow the window was on your part: how little had to change — a fraction of the section, a slightly different fit-up, the second one out of the fixture — before the result stopped being the same result.
A process you can only hit when everything is right isn't a process. It's a run of luck, and it ends on the Tuesday your best operator is off.
What you get instead: the edges of that window, found on your own part, so walking away is a decision you made with evidence rather than a supplier's opinion.
Answer four · This one we won't take
The other ending is shorter.
Sometimes it's what you told us about the part, sometimes it's what the first weld did — either way it becomes clear this isn't work we should be taking, and you're told which limit your part landed in rather than being handed a machine anyway with a note about careful operation.
You're also shown where that limit is written down, so it can be checked against somebody else's judgement rather than taken on ours. A limit you can check is worth something to you even if nothing else about this trip was.
And you aren't left holding the limit on its own.
The same reply names the change that would work on your part instead — not a list of things that sometimes work, but the one that applies to the piece we had in our hands, described closely enough that you could hand it to whoever would have to make it.
Whether it's worth making is yours to weigh. What you no longer have to do is guess at what it would be.
What you get instead: a named limit you can take to another supplier, and the one change that would work on your part.
And the part itself: it stays yours either way — back to you if you ask for it, left with us if you'd rather it didn't travel again. Which of those happens is your call, not ours.
Which Part to Send — Usually Not the One That Brought You Here
Three candidates sit on most benches. Two of them get you a precise answer about the wrong thing.
The one that failed. It's the reason you're reading this, and it's the worst sample you own. Whatever wrecked it is now part of the evidence — the crack, the burn-through, the section that was already thin — and the answer comes back describing that history instead of your work. Put it in the box as background if you like. Don't send it as the sample.
The clean one you'd be happy to photograph. Deburred, polished, fixture marks taken off, sometimes with the neighbouring component removed so it travels safely. Every one of those courtesies deletes a condition we were meant to judge, and the answer comes back about the tidy version rather than the one you'll actually be running.
The one you'll be running. Same alloy, same section, the neighbour still attached, and in the state it will be in at the moment the weld happens — before the coating if that's when you'd weld it, straight off the casting if that's when you'd weld it. Ugly is fine. Representative is the entire requirement.
One thing to settle while you're still choosing. Your part changes on this trip: it gets a weld on it, and that isn't something you undo. If that matters — a piece you were going to sell, something that belongs to your own customer, a part you need on the line next week — the answer is to pick a different one, not to send that one and hope.
When to send two. One or two is the range — two when the same joint runs across a condition you can't collapse into one part: two suppliers for the same alloy, or the same piece before and after the step you're unsure about. Past two we're welding your batch rather than answering your question.
If you still have to go and get one, take it from the middle of a run. The first one off and the one in the reject bin are unrepresentative in the same way, for opposite reasons.
What to Say About It — Six Lines That Change the Answer
Paste this block into your email and fill it in. Each line is here because it changes what we'd do; nothing else is.
- Alloy, and who supplied it. — Two batches of the same alloy weld differently often enough that we'd rather know than assume.
- What's already been done to it. — Plating, heat treatment, work hardening: each one changes how the surface takes the first pulse.
- What sits next to the joint, and how close. — Resin, cement, a seal, a set stone, a component that can't be removed. This decides more settings than the metal does.
- Whether that joint has been welded, brazed or soldered before. — Reworking an existing joint is a different job from making a new one, and it's the line people leave out.
- What happens to the part after this step. — Polishing, plating, potting or assembly downstream can rule out an answer that would otherwise be fine.
- What "good" has to mean here. — Cosmetic on a face that shows, structural under load, or sealed against leaks: three different welds on one joint.
You know which part, and you know what to say about it.
That's the whole preparation. Everything after this is about what happens once it's in the post.
How the Part Travels — Out to Us, and Back if You Ask for It
The unglamorous half. Written down because it's the half people stall on.
Going out. Pack it as small as it deserves — for most of this work a padded envelope does it — and write the note before you seal the packet, because the note is the first thing that gets read. The packet goes to Layer 1-2, Building B8, 28th Street Leisure Business Center, Panlongcheng Economic Development Zone, Huangpi District, Wuhan, Hubei Province, China.
While it's here. How long before you hear back: within two working days of the part arriving. The bench runs Monday to Saturday, China time, so the only day that doesn't count is Sunday. What arrives then is the answer itself — the weld attempted and what it told us — not a note saying the packet turned up.
Coming back. The sending and the return are both arranged in our first reply to you, before anything is boxed — so no part of it is something you find out about at the end.
What Sending a Part Signs You Up For
The questions that get asked by email before a part goes out. Answered here so they don't have to be.
Does sending a part mean I've started buying something?
No. There's no order behind it, nothing to cancel, and no point at which having sent a part turns into having agreed to anything. The evaluation itself is free — the welding, the answer, the parameter sheet. What the sending and the return involve is in that first reply.
Who gets in touch afterwards, and how often?
One person, once, about your part: what we found and what we'd do. If the answer was no, that's where it stops unless you write back. There is no campaign running behind this.
Can I do this without a company name on it?
Yes. A name and an email address is the whole requirement. Plenty of the benches we answer for don't have a letterhead, and none of them get asked for one.
I'm doing this for a company that hasn't decided anything yet. Is that a problem?
It's the normal case. Nothing you send commits your employer, and nothing that comes back is addressed to a purchasing department. If what would actually be useful to you is something that survives a meeting you're not chairing, say so and it gets written that way.
And if we go quiet after that?
Then it's quiet. The part got welded, the question got answered, and there's nothing outstanding on either side.
Nothing above this line has been about our machines.
Send One Part for a Free Weld Sample Evaluation
One part, the six lines from above, and somewhere to send the answer.
This goes to Elowen, our micro welding sales engineer. The reply tells you who to mark the packet for and what to write on it.
That's with Elowen.
You'll have a reply within 24 hours on business days (Mon–Sat, 9:00–18:00 GMT+8), and it will tell you who to mark the packet for.
That didn't go through.
Email [email protected] or message +86 15327155363 and we'll pick it up from there.
[email protected] · WhatsApp +86 15327155363 · Mon–Sat 9:00–18:00 GMT+8 · Elowen, Micro Welding Sales Engineer