A two-part epoxy AB adhesive fails in a way that is almost designed to defeat incoming inspection. Mix it slightly off ratio and it still gels, still goes hard, still passes the pull test your line operator does on the third unit of the shift. It ships. Then, somewhere between eight and sixteen weeks later, drivers start coming back from a humid market with a rub, a buzz, or a spider that has crept loose from the frame — and by then you have built forty thousand more of them the same way.
That gap between what the bond looks like on day one and what it is on day ninety is the whole argument for treating mix ratio, pot life and fixture time as procurement data rather than chemistry trivia. Aosibo Adhesives, a Dongguan supplier whose core range is loudspeaker assembly adhesives, also lists two-part epoxy (AB) and acrylic (AB) systems alongside cyanoacrylates for general industrial bonding — and states plainly that mix ratio, pot life, fixture time, full-cure schedule and substrate compatibility are all order-specific. That is an honest position. It is also a blank you have to fill in yourself, in writing, before a purchase order means anything.
The four clocks, and why buyers confuse them
Most quote conversations collapse four different time values into one word: "cure". They are not interchangeable, they are measured under different conditions, and only one of them sets your line speed.
| Clock | What it measures | Where it bites you | Condition it must be quoted with |
|---|---|---|---|
| Pot life (working life) | How long the mixed A+B stays low enough in viscosity to dispense | Batch size at the mixing station; how much you throw away per shift | Mixed mass and temperature — a 100 g cup at 25 °C is the common lab basis |
| Open time | How long the applied bead stays wet enough to accept the second substrate | Operator dwell between dispensing and mating parts | Bead geometry, temperature, humidity |
| Fixture time (handling strength) | When the joint holds itself well enough to move to the next station | Work-in-progress buffer, rack count, floor space | Temperature, substrate pair, and the strength threshold used to define "fixtured" |
| Full cure | When mechanical and thermal properties reach their published values | When you are allowed to run final QC and pack | Temperature and dwell; often quoted at both room temperature and an oven schedule |
Two practical consequences fall straight out of that table. First, pot life is a property of a mass, not of a chemistry. A 100 g cup of mixed amine-cured epoxy generates its own heat, which accelerates the reaction, which generates more heat — so the cup can go from workable to a smoking solid brick far faster than the same material laid down as a 2 g bead on a top plate. Buyers who dose from a large mixed batch and buyers who dose from a static-mix nozzle are living under two completely different pot lives, and only one of them is on the datasheet.
Second, "fixture time" only means something once someone names the strength threshold behind it. A common industry convention defines the fixture point at a nominal handling strength — figures around 0.1 MPa are widely cited — but suppliers do not use one universal number, and a supplier quoting a shorter fixture time may simply be using a looser threshold. Ask what strength value defines fixture, on which substrate pair, at what temperature. If the answer is a shrug, the number is not comparable to anyone else's number.
Mix ratio is the one specification people get wrong
Here is the failure that shows up more than any other on lines running AB systems bought from a new supplier: the ratio is stated on one basis and metered on another.
A 2:1 mix ratio by weight and a 2:1 mix ratio by volume are not the same mix. Epoxy resins and their hardeners rarely share a density; a resin around 1.15 g/cm³ paired with an amine hardener nearer 0.95 g/cm³ means a volumetric 2:1 delivers something closer to 2.4:1 by weight. If the datasheet quotes weight and your dispensing equipment meters volume, you are off ratio on day one and no one on the floor did anything wrong.
Off ratio is not a linear penalty either. Both directions hurt, and they hurt differently:
- Excess resin (hardener-starved). Unreacted epoxide groups stay in the network. Crosslink density falls, glass transition temperature falls, and the bond becomes progressively more sensitive to heat and moisture. It will often still feel hard. It is the version that passes the day-one test and loses shear strength through a humid summer.
- Excess hardener (amine-rich). Free amine plasticises the matrix and stays mobile. In humid air, surface amine reacts with atmospheric carbon dioxide and water to leave a greasy carbamate bloom — the "amine blush" that shows up as a waxy film and, on a subsequent bond, as an adhesion failure nobody can explain. In a Dongguan summer, or in your own plant in July, this is not a rare edge case.
The insider detail worth carrying into the quote conversation: an off-ratio epoxy frequently develops higher initial hardness than a correctly mixed one, because the excess component acts as a filler before it acts as a defect. Line QC that measures hardness or does a quick destructive pull is therefore biased toward passing exactly the batches you most want to catch. If ratio control is the risk, the test that finds it is a humidity-aged or heat-aged sample set, not a same-shift pull.
Cartridge, static mixer, or separate A and B containers
How the material arrives changes your cost per bonded joint far more than the price per kilogram does. The source data for Aosibo states MOQ as 1,000 units without defining whether a unit is a bottle, a cartridge, a syringe or a kilogram — and for adhesives those differ by orders of magnitude in both value and consumption. Settle the supply format before you compare any two quotes.
| Supply format | What you buy | Line equipment needed | Where the hidden cost sits | Best fit |
|---|---|---|---|---|
| Dual cartridge + disposable static-mix nozzle | Pre-metered A and B in one cartridge | Manual or pneumatic cartridge gun | Purge volume discarded each time a nozzle is fitted; nozzle unit cost; cartridge freight per gram of adhesive | Low-to-medium volume, frequent changeovers, mixed-model lines |
| Separate A and B containers, hand-mixed batches | Bulk pails or bottles of each part | Scale, mixing vessels, timer, disciplined SOP | Operator ratio error; batch waste when pot life expires mid-shift | Low volume, large-mass bonds, repair and MRO |
| Separate A and B feeding a meter-mix-dispense unit | Bulk of each part | Metering pumps, ratio check routine, calibration schedule | Capital cost and a real calibration discipline; ratio drift as pumps wear | High volume, one or two adhesives, stable takt |
Three things that never appear on a quotation and always appear on your consumption report:
- Purge waste. Every time a static-mix nozzle goes on, the first few millilitres out of the tip are off-ratio and must be discarded. Multiply that by nozzle changes per shift — after breaks, after any stoppage longer than the pot life, after a model changeover — and a line running frequent changeovers can throw away a meaningful share of every cartridge. Budget it, and ask the supplier for the recommended purge volume for the nozzle they specify.
- Nozzle element count. Static mixers are specified by the number of mixing elements and bore. A short nozzle on a high-ratio system (say 10:1) does not mix properly, and you get a striped, partially cured bead that looks fine and behaves like an off-ratio mix. If a supplier offers cartridges, ask which nozzle part number they validated the mix ratio with.
- Remaining shelf life on arrival. Two-part systems and cyanoacrylates age from the date of manufacture, not the date of shipment. Ask for shelf life from date of manufacture and put a minimum remaining shelf life on arrival into the contract as a term — this is one of the few adhesive risks you can eliminate with a sentence rather than a test.
If you are structuring the whole enquiry rather than just this one line item, the discipline in how to write a product spec sheet for a Chinese factory applies directly: the spec you send defines the quote you get back, and adhesive quotes diverge for the reasons set out in why China quotes for the same spec differ.
What a mis-metered mix does three months later
Trace the field-return signature backwards and it is remarkably consistent.
On a driver, an under-cured or off-ratio bond does not usually let go all at once. It creeps. The magnet-circuit stack shifts by a fraction of a millimetre and the gap geometry changes — see the separate discussion of magnet-circuit adhesive on the top plate and yoke for what that joint is carrying. A centring bond softens above its now-reduced glass transition temperature during a long high-power session and the coil starts to rub. A spider-to-frame joint with residual free amine picks up moisture and loses shear strength progressively rather than suddenly. All of these read to your customer as "buzz" or "distortion at volume", and all of them arrive as a rolling return rate rather than a single lot recall — which is precisely why they cost so much to diagnose.
The tell that separates ratio error from every other cause: the failures are not confined to one production date. Genuine bad-lot problems cluster; metering problems smear across whatever period the equipment or the SOP was wrong, and often affect only the shifts where a particular operator or a particular station was running.
The counter-measure is not a better adhesive. It is a metering record. Keep a per-shift log of the ratio check — a mass check on a dispensed shot into a cup, weighed A-only, B-only and mixed — and retain a small cured witness sample from each shift for aged testing. When a return wave arrives, you have a population to test against rather than a theory.
What to ask before you quote
Aosibo's own instruction to buyers is the right shape: send the two substrates and the cure window you have available. Most buyers do the opposite — they describe the bond and ask what cure it needs, then discover the answer does not fit the equipment they own. Lead with the constraint.
Before you accept any AB system into a driver line, get these in writing, per grade:
- Mix ratio, and the basis — by weight or by volume — plus the density of each part so you can convert.
- Pot life, with the mixed mass and temperature it was measured at.
- Open time under conditions resembling your plant, not a 23 °C / 50 % RH lab.
- Fixture time, with the strength threshold and substrate pair that define it.
- Full-cure schedule at room temperature and at an oven schedule, with the dwell and temperature for each.
- Substrate compatibility for both of your actual substrates, named — not "most plastics and metals".
- Supply format: cartridge, syringe, bottle or pail; net contents; and if cartridge, the validated nozzle part number and recommended purge volume.
- Shelf life from date of manufacture, storage temperature, and an agreed minimum remaining shelf life on arrival.
- What "1,000 units" means on the MOQ line — bottles, cartridges, syringes or kilograms — before you compare the price to anyone else's.
Aosibo's stated terms give sampling at 7 days and production at 15 days, both company-stated and both worth confirming in writing. Use the sample cycle to run the ratio and ageing tests above rather than a same-day pull test; that is the difference between a sample that proves the material and a sample that proves the courier works. The company's factory profile on ChinaMakersHub lists the loudspeaker range these AB systems sit alongside, including the centring adhesive offered in one-part, epoxy and halogen-free versions, which is often the more natural starting point if the bond in question is inside the driver rather than on it.
Common questions
Can I substitute a two-part epoxy for a one-part adhesive to shorten cure time?
Sometimes, but the trade is rarely free. A two-part system removes the oven dependency and can fixture at room temperature, which frees oven capacity — but it adds a metering step, a pot life constraint, waste, and a new failure mode your incoming inspection is not currently set up to catch. If the driver is oven-bound today, the honest comparison is line seconds and rack space against metering discipline and scrap, not one cure time against another.
Why does the same adhesive have a different pot life at my plant than in the supplier's lab?
Temperature and mass. Reaction rate for most amine-cured epoxies rises steeply with temperature — a rough working assumption is that rate roughly doubles for each 10 °C increase — so a pot life quoted at 25 °C can be dramatically shorter on a 33 °C shop floor in summer. Mixed mass matters just as much, because a larger mass retains its own reaction heat. Ask for the mass and temperature basis, then adjust for your conditions rather than assuming the number transfers.
How do I check mix ratio on the line without laboratory equipment?
A bench scale with 0.01 g resolution covers most of it. Dispense a shot into a tared cup, record the mass; do the same with each part dispensed alone through a purge; compare against the specified ratio on the correct basis. Do it at the start of each shift and after any equipment intervention, and keep the log. Pair it with a retained cured witness sample per shift, stored for later aged testing — the log tells you what the ratio was, and the witness sample tells you what the ratio produced.