Ted Mooney, P.E. RET
- Pine Beach, NJ
The authoritative public forum
for Metal Finishing since 1989
-----
Test to determine if powder coating is fully cured
Please see also --
• Topic 37637 "Re-curing powder coat"
• Topic 15333 "MEK/Solvent Rub Test? Cure Tests"
• Topic 46160 "Can we do powder coat on top of powder for rework"
Q. We are mainly powder coating galvanized tubes that get welded into different geometries. Flat steel tabs and other components can also be welded onto the tubes. We also do some straight 5" OD aluminum tubes.
I am trying to determine if we are adequately curing our product.
What is the best way to determine this?
We currently do cross hatch testing but we never seem to fail.
I have my doubts that we are 'fully' cured, especially on ALL of our parts considering the differences in geometry- sizes and shapes.
Engineer - Minnesota
July 22, 2009
A. Hi Matt
The other way of determining the curing effect is "Solvent rubbing resistance"
Solvent used for this test is normally Methyl Ethyl Ketone.
Test Method:
Dip a plain cloth in the described solvent and make 10 double rub on the part.
Evaluation:
If your part is not cured properly the cloth will have some paint on it.
Hope this test will helps to clear your doubts.
Let me know if you have further assistance on this.
Vijendran
- Chennai, India
|
A. We use epoxy powder. Rubbing with a swab and acetone ⇦this on eBay or Amazon [affil links] Warning! highly Flammable! will show if it is fully cured. David Makinson- West Union, SC, USA July 28, 2009 A. Cross hatch test isn't really a test for cure, rather for adhesion. Its possible to have good adhesion without full cure, so this test isn't telling you much about cure.  Geoff Crowley Crithwood Ltd. Westfield, Scotland, UK  July 29, 2009 Rule of thumb is to check the thickest material, if the thickest material is cured the whole part should be cured.  Sheldon Taylor supply chain electronics Wake Forest, North Carolina July 29, 2009 A. As far as I am aware the cross hatch method you currently use is a test for inter coat adhesion. A more reliable test would be to gouge a short groove in the powder film using a silver coin; this may take some pressure to get down to the metal substrate but well worth the broken wrist :-). Birmingham, United Kingdom July 29, 2009 |
Cured powder coating dissolving during solvent test
Q. Hi everybody,
I want to share a powder coating problem we have recently had. We did powder coat some electrogalvanized steel sheets but after curing some parts of the sheets had very bad cissing/dewetting defects. We applied some solvent (acetone) on one of the pieces in order to peel away a little portion of the softened coating and look at the surface of the sheet, but the drop of solvent instantly dissolved the coating; the metal underneath was visible after wiping the surface with blotting paper. The same thing happened on most of the other sheets that did not exhibit any defect with both MEK or acetone, so it seemed that no crosslinking took place.
The powder supplier took some pieces in order to perform a TGA analysis but it turned out that the powder was fully cured. Has anyone every had the same experience?
Powder coater - Berlin, Germany
April 21, 2016
A. Are you using a convection oven or IR?
If IR then the problem is easily explained.
Regards,
Bill
Trainer - Salamander Bay, Australia
April 26, 2016
Q. Hi Bill,
We are using a convection oven.Could cross-contamination between different powders lead to such defects?
Thanks
- Berlin, Germany
April 28, 2016
A. Hi Again,
Have you, yourself conducted an MEK 32 rub cure test on the offending components?
If not, please do so.
Cure testing with polyesters has become a bit of a black art.
The long established and proven 32 rub test with MEK has been somewhat hi-jacked by some, most notably Orika/Dulux who tend to substitute the solvent resistance test as an alternative.
This they do by providing the customer with "cure test solvents" which are not straight MEK.
In general to your last post I would say never, ever mix powders from different supply.
There are very good reasons why every box is marked with both batch & box numbers.
This is largely to provide traceability in the event of problems such as you are encountering.
Hope this helps,
Regards,
Bill
Trainer - Salamander Bay, australia
Q. Hi Bill,
Of course when I say cross contamination I mean accidental and minimal contamination after colour change, not that we're intentionally mixing powders from different boxes. Moreover, I always perform a solvent rub test on both the new sample and on a fully cured one (in this case provided by the supplier), I know that every powder can have a different behaviour. The thing is, I don't even need to rub; the powder dissolves instantaneously as a drop of MEK or acetone is put on it. This usually means that the powder isn't properly cured, but we're more than sure that the curing cycle was appropriate and the Tg of the coating proves it.
Regards,
Anna
- Berlin, Germany
May 2, 2016
Hi Anna,
I am a little confused by your posts.
I originally understood that you had conducted a 32 rub MEK cure test with satisfactory result and then were disturbed to find that a droplet of Acetone completely dissolved the powder film.
Your subsequent post seems to contradict that position?
I am assuming that you are using Polyester Powder?
If so, at the end of the day your product is failing the standard 32 rub MEK cure test?
If that is the case then you only have two variables:
1 Faulty batch of Powder(unusual but not unheard of)
2 Faulty cure schedule (time at metal temperature)
The most common error with cure schedule is to assume that oven air temperature translates directly to metal temperature. Have you run a thermal data logger through your system attached to a piece of the same material that is creating the problem?
If not it is essential that you do so to create a true time/metal temperature graph (cure schedule)
I have some difficulty with your references to glass transition temperature.
I may be wrong but I can conceive that initial Gelling of the powdercoat could be regarded as Tg.
Initial gelling does not ensure that full cross polymerisation of the powder has occurred, rather that it has merely initiated.
Your Powder Supplier must have a thermal data logger if you do not. They should provide a check against your own methods on a regular basis in any instance.
Please keep me updated with your progress.
Regards,
Bill
Trainer - Salamander Bay, Australia
May 4, 2016
Q. Hi Bill,
I'm sorry for the confusion. In my first post I said that some of the pieces showed some cissing/dewetting defects and other were free of surface defects. The test with MEK and acetone instead gave the same result both on the pieces with and without surface defects (total dissolution). There were anyway a few pieces at the end of the order that passed the solvent test without dissolving and that didn't show any surface defects. We used a new box of powder for the whole order, a polyester one.
The Tg (glass transition temperature) increases with the crosslinking of the powder. One of the methods to verify the curing of a sample is to measure the Tg of a fully cured sample and compare it with the Tg of the investigated sample. In this case the Tg were similar, and this should mean that the pieces are properly cured. We cannot then understand why the coating dissolves in mek/acetone if it is cured.
The cure schedule is under control, we do have a thermal data logger.
We thought that the problem could be related to a faulty batch of powder, but as I said a few pieces did pass the cure test and the supplier tells that there can't be differences in powder composition inside the same box.
Thanks,
Anna
- Berlin, Germany
Hi again Anna,
I am intrigued.
Your Powder supplier is probably correct about consistency within the same box.
It seems the most likely way to achieve inconsistency in the powder applied would be to do so within the process.
If you are using a vibratory feed system rather than an aeration system for example (this usually only causes problems with metallic powders).
I have not asked if your process is a batch system or a fully conveyorised system?
I recall training a group in Australia who had a batch system. Their oven was set up with a system which kicked in to time once the object had reached cure temperature and then send an alarm to unload.
They always had perfect Time/ Temperature graphs.
Unfortunately, they also had consistently undercured products.
Sound familiar?
Their problem was that no-one understood the significance of where to place the thermal data logger clamps.
For convenience they left it clamped to the inner surface of the oven rather than the workpiece.
The only cured pieces out of that oven were the last ones.
I hope that I have assisted you and would very much like to hear how your problem is resolved.
Best Regards,
Bill
Trainer - Salamander Bay
A. Hi Anna,
Just check whether you are using right temperature grade of grease or lubrication oil in your conveyor.
A relatively low temperature grease (very close to the curing temperature, say 250 °C) will burn and form fumes. These fumes can interfere your coating randomly; however more temperature you cure the substrate. Watch for smoke from oven entry/exit doors.
Use lube oil with a working temperature of 350 °C and avoid greasing, if used.
Please update your feedback.
- Chennai, India
July 1, 2016
Solvent Resistance Test for Hybrid Powder Coat Epoxy- Polyester
Q. I've done some cure tests with acetone on a new hybrid epoxy polyester powder and they tend to instantly remove gloss and change the surface. There is significant coating dissolved after 20 rubs. Online research hasn't given much clear information about hybrid chemical resistance. The powder vendor's data sheet says that acetone will dull and soften at 1 hour. Can I assume that this means that my parts have failed the cure test?
Terry MillerEngineer/Project Manager - Huizhou, Guangdong, China
November 21, 2016
A. We in Australia, refer to this Epoxy/Polyester blend as Hybrid. Normal cure test procedures are used: 32 rubs with MEK resulting in negligible discolouration on the white rub rag signifying acceptable cure. We also have a solvent resistance test (quite separate) which uses Acetone.
It appears that your product is significantly under-cured.
The usual error for "young players" is to assume that oven air temperature directly translates to metal cure temperature.
This is obviously quite wrong.
I would suggest you acquire a thermal data logger and check the actual "cure schedule" : time at metal temperature you are actually working with.
In the first instance you could impose on your powder supplier to conduct a test with their equipment.
The logger should cost <US$2000.00.
It is essential equipment for any quality coater.
Hope this helps.
Regards,
Bill
Trainer - Newcastle Australia
Q, A, or Comment on THIS thread -or- Start a NEW Thread