Blogs & News

Copying is the easy road.

We have built bakery machines in this trade for over twenty years. Our founder and chief engineer, Robert, comes from food science and engineering himself. We design our own machines, cut our own molds, assemble them one by one — and we have taken apart plenty of competitors' machines.

Strip enough of them down and one thing becomes clear: the three most valuable numbers on a machine are not on the drawing.

This article is not here to blame anyone. Copying in this trade is more innocent than you think, and messier than you think. Let me do the math: what copiers save, what they miss, and how you — the person buying the machine — can tell over one lunch whether a supplier really has R&D capability.

1. A Fair Word First: Copying Is Often Not a Morality Problem

Have you ever walked through an equipment factory?

Lathes turning, milling machines humming, five or six fitters building machines on the line. Big floor, full equipment, looks strong.

That is real strength. But it is manufacturing strength — not necessarily R&D strength.

Ask one question: why is the sealing face on this machine 0.04 mm? Why not 0.08 mm, and why not 0.02 mm?

At most factories, nobody can answer that.

Not because they refuse to. Because nobody at their factory has ever asked. That drawing was not theirs.

We look at a factory's capability in three layers:

The first two layers come with money. The third does not — it needs people, time, and a system that allows failure.

Most factories have the first two capabilities. The third is not something a few machines and a few new hires will give you.

2. Why Copying Pays: A Set of Numbers You Can Actually Add Up

The copying route is short.

Buy a sample machine, tear it down, measure it, draw it, source the machining, assemble, sell. A few months if you move fast. Risk is near zero — the machine has already been validated by the market somewhere else.

The R&D route is long.

Set up the project, work out the principle, prototype, try and fail, run bench tests, cut molds, run small-batch production, validate — and only then dare to sell. Counted in years. Every step can fail, and the money and time lost stays with you.

So the difference is not courage. It is how the math is kept —

• Copying: you keep the upside, someone else keeps the risk — they already paid the tuition for you;
• R&D: the upside is uncertain, and the risk is all yours.

At a factory judged on this year's shipments and this month's cash collection, "no innovation" is not laziness. It is rational.

But a copied machine has three parameters that never line up.

Those three numbers are not on the drawing, and not on any quotation.

3. You Can Copy the Shape, Not the "Why"

Whether a Dough Divider & Rounder hits its piece weight comes down to air tightness.

Vacuum suction: hold the dough, then you can divide it accurately. Air tightness depends on three places: the base, the pusher, and that sealing face up front. All three rest on machining precision.

The first number is 0.04 mm.

We hold our sealing face to 0.04 mm. Leading brands generally sit at 0.08 mm, and the industry norm is 0.10 mm.

That is a one to one-and-a-half times gap. The eye cannot see it. The dough can feel it. Marked 110g, competitors hold only 95g in steady production; we mark 110g and we give 110g. (110g is the capacity of the 3P standard version BAFP-3P divider, 25–110g per piece.)

The second number is 185 mm.

Contact face width. Leading brands run 147 mm. When your dough is on the stiff side, that 38 mm decides whether the dough goes in and whether the vacuum holds.

The third number is the special machining on the cone surface and the side ring.

It gives one result: full-surface rounding with no extended track. Competitors have to extend the track, and the added cost lands on the customer; with the machine fixed, the customer has to change the formula to accommodate the machine. We run the order the other way around — the machine adapts to your dough, and your dough gets rounded on its full surface.

The Smart Dough Divider & Rounder (Vertical) (model BK-Smart) is a separate product line, so its numbers have to be stated separately: piece weight tolerance 0~3g, where ordinary semi-automatic machines in the industry are ≤5g; smallest piece 10g, where other models start at 20g; cutter disc changed in 30 seconds, where the old way means stripping the whole machine and re-adjusting on reassembly; oil-free drive, with the oil seal gone, so there is no such thing as an oil leak; industrial linkage switch saves 50% of the power on standby, up to about 825 kWh a year; noise cut by as much as 80%.

The copier takes those dimensions away. But he measures 0.04 mm without knowing why 0.04 mm, how to inspect it, or what to do when it fails inspection.

You can copy the dimensions. You cannot copy why the dimension is what it is.

The "why" from the design process stays in our factory. It came from years of trial and error, and it stops being worth anything the moment it is written into an article.

As for what exactly we changed on these two machines, we wrote about it separately: Twenty Years Without a Change From Competitors — Why Did We Change More Than Twenty Things?

4. It Is Not That They Refuse to Build an R&D Team — the System Cannot Carry It

A team that can set its own drawings needs at least seven things:

• Five disciplines: mechanical, electrical control, software, process, testing;
• A budget that allows failure — prototypes, molds and scrap all cost money;
• A workshop that can build its own trial units, instead of waiting on outside contractors every time;
• Test benches and inspection methods for validation;
• Mold makers and suppliers willing to change along with you;
• Test data that flows back onto the drawing;
• A rule: admitting "this version does not work" is progress, not an incident.

Now look at the reality in most factories.

The owner came up through sales or machining; R&D = whatever the technician does on the side; drawing = tear down a sample; KPI = this month's shipments; failure = waste.

So people do not come, do not stay, and cannot get going. A real structural engineer joins, and six months later finds that nobody discusses load paths with him, nobody can validate his idea, and his proposal waits three months for an outside contractor. He leaves. Whoever stays keeps copying — because copying is the only way to hit the deadline.

Our route grew slowly.

Our team includes engineers who have spent over twenty years in precision machining, and an engineering group with more than a decade behind it. We design our own machines, cut our own molds, assemble every unit ourselves. The Tapered Divider & Rounder covers its weight range with 12~13 sets of molds, and behind those molds sit several hundred sets of validation — not drawn on paper, but cut one set at a time and tested batch after batch of dough. The Smart model reaching 10g fine division rests on three things holding at once: in-house mold making to control tolerance, precision assembly to control clearance, and stable output at volume.

We redid the main control board: high-voltage wiring stays off the PCB, high and low voltage are physically separated, and every unit is pressure-tested before it leaves. When a machine faults, often a whole board swap is enough — no waiting for an electrician to trace wires one by one.

We also keep a dumber rule in the plant: run every machine empty first, then run it again with about 2kg of dough.

A few things are still in progress — the feel and the effort of the pressing mechanism, and how the head flips for cleaning. They are all tested in our own workshop, and we do not talk about them until they are ready.

R&D is not one person's job. It is a whole system holding it up.

5. Who Ends Up Paying: A Copied Machine Shows Its Hand Two Years Later

A copied machine does not break on day one.

For the first three months it behaves normally, and can even look impressive. The difference shows up two years later.

Piece weights start to drift, and the answer is "that is your dough".

Sand holes in cast iron get eaten by sugar and salt, air tightness drops, and the weight ceiling drops with it. Locating parts on bronze bushings wear looser and loosen accuracy along the way; use a bearing seat instead and one part swap puts it back to factory condition.

The fine mesh on the flour duster sheds iron grit — one customer was fined over it; switch to a food-grade plastic mesh plate and nothing sheds, plus it can be replaced on its own.

Wear parts in plastic are cheap; anodised aluminium parts cost far more. What you are paying for is not the part price — it is whether you have to strip the machine down in two years.

When dough-contact surfaces cannot be cleaned thoroughly, the workshop rule becomes "skip it if you can".

The most expensive item, in fact, is not a part at all.

It is downtime. A factory with no R&D capability can only wait when it meets a fault it has never seen; we have our own machining equipment and engineers with over twenty years in precision machining, and we give you a plan within 24 hours.

Many customers buy someone else's machine first, and only then come back to us.

We also wrote separately about materials and contamination: Why Do We Research So Much? Because These Two Machines Have No Answers on the Drawing

Cheap when you buy it. Expensive when you run it: downtime, rework, staff turnover.

6. Eight Questions to Test Whether a Supplier Has R&D Capability

You do not need to know machinery.

Take this list and ask. Once you are done, you will know:

1. If you need pieces above 100g → ask them to show the weight distribution of a continuous run of 100 pieces on the spot, not just one piece?
2. If steady piece weight matters → ask how fine the sealing face is, and how they inspect it?
3. If your dough runs wet or stiff → ask "do I need to change my formula". If they want you to change the formula, they are passing the cost to you?
4. If you want to know whether the machine is a copy → ask whether this structure is theirs, or whether the industry has always done it this way?
5. If you worry the test run is theatre → ask who pays for the trial, who shows up, and what happens if it fails?
6. If you worry about line stoppage → ask whether wear parts are in stock, and how fast they can give you a plan when something breaks?
7. If you worry that no one owns the drawing → ask who drew it, what changed in the last version, and why?
8. If you clean the chamber once a day → ask whether the strip-down works by feel, whether hands stay in the stroke, and whether there is any risk to the operator?

I do not suggest you use these questions to corner any supplier.

You are there to check one fact: whoever can answer them has R&D.

Copy one machine: a few months. Develop one: counted in years.

I understand why the industry copies. But someone pays that bill, and it is the person buying the machine.

Our way is plain: we draw our own drawings, cut our own molds, run our own tests, make our own changes. If you have a formula and dough, we will run the test.

Message Robert on WhatsApp and we will send you a test-run video with the same dough.

No guessing. Give us your formula, we run it on site, and we ship once it performs. Stiff dough is fine too — we have run bagel dough before.

Scope statement: This article applies to the Smart Dough Divider & Rounder (Vertical) (model BK-Smart), the Semi-Automatic Dough Divider & Rounder (Vertical), and the Tapered Divider & Rounder (divider + tapered rounder). The Smart Dough Divider & Rounder (Vertical) and the Semi-Automatic Dough Divider & Rounder (Vertical) are two machines with different functions, not two names for one machine, so their figures do not transfer between them. The 0.04 mm, 185 mm and 110g figures belong to the Tapered Divider & Rounder, of which the 110g figure is the per-piece capacity of the 3P standard version BAFP-3P divider. The material claim for the Tapered Divider & Rounder refers only to "the dough-contact die head assembly being 304 stainless steel". Final data is set by the actual quotation and contract; contact official customer service if anything is unclear.