HDT testing. Methods for workshop

Discussion in 'Fiberglass and Composite Boat Building' started by viz, May 31, 2009.

  1. viz
    Joined: May 2009
    Posts: 18
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Auckland NZ

    viz Junior Member

    G'day,
    I confess ours is a sailplane project rather than a sailboat. I hope someone can help with some ideas. I'm looking for a simplified method of checking whether our epoxy resin samples are over a target HDT. This is just to confirm our final cure state after postcure. I guess we could also use it to check for the initial cure state also.

    I read a little on formal definition of HDT the test.
    http://www.ttc.bayermaterialscience.com/bpo/bpo_ttc.nsf/0/3C1C1194910DC464C1256D59002CA58E.

    The proper procedure is not really practical in the small workshop. I wonder what others have done. I can stabilise the temperature in an electric kettle (jug) with water and warm our resin samples then try bending them with pliers. We did this. But how much force should it take? So is there some easy idea for this. Some data below.
    Resin ADR 246
    Hardener ADH 25, 28, 185 sometimes blended
    HDT ultimate 94, 100, 88degC
    Resin samples diameter=30mm, thickness=5mm.

    Cheers
    Gregg
     
  2. Jimbo1490
    Joined: Jun 2005
    Posts: 785
    Likes: 41, Points: 28, Legacy Rep: 527
    Location: Orlando, FL

    Jimbo1490 Senior Member

    Testing HDT is not all that hard, but in order to make the results meaningful you'll want to be able to compare your test results with a published result from, for instance, manufacturer's data sheets, in which case you'll need to use the exact same test procedure that the manufacturer used, like ASTM D648, which is a commonly used HDT test regimen. In this case the test method will determine the coupon size. Fabricate the coupons accurately and do multiple tests on many coupons cured under the same conditions. Keep track of the batches; mark everything!

    The trick is going to be to use accurately sized test coupons. You can make a small mold so that the thickness of the coupons can be repeated. Make them a bit too big and then trim them down to exact size with fine sandpaper on a nice flat surface like a drill press table. This will also eliminate things like the edge meniscus which could throw off the test by changing the cross sectional shape of the coupons.

    ASTM D648 called for the sample to be submerged in a liquid, IIRC. The temperature of the liquid gets raised at a certain rate while a deflection force is applied. When the coupon deflects the specified amount, the temperature is noted as the HDT for the material at the test parameters.

    You can make your own low-cost immersion bath from a process controller and one or more fish tank style heaters, with Ni-Cr wire coils encased in glass immersed in the liquid. The coils would be under the control of the process controller which is receiving real time temp. data from a thermocouple or an RTD. Any modern controller will let you program a rate of temp. increase that's slow enough to let the entire thickness of the coupon stabilize at the new temp. Also the the controller will show the process temp. in its display screen so that you can note the temp at the moment the deflection threshold is reached. Get a controller with "PID and fuzzy logic" as it will self-program to work with your setup (bath size, heater size, etc.) so you will not have to spend any time learning to program it. You should be able to buy everything you need for around $300-$400 buying new. If you scavenge it all off of ebay (there's TONS of controllers, thermocouples, TC wire & plugs, RTD's and etc on ebay) there's no telling, but I would say you can get it all for $100 or so.

    If you guys are avid DIY people, then building the bath and molding coupons should not be too big a project, especially if $$$ is tight and you can't afford the services of a test lab right now (or ever:( )

    Jimbo
     
  3. viz
    Joined: May 2009
    Posts: 18
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Auckland NZ

    viz Junior Member

    Thanks Jimbo,
    I think you're right, we could just pull finger and do a stock ASTM D648 test for future work. The glider prototype we are building is mostly built and so the resin coupons are already existing at the wrong size. At at the moment we is just me and with a serious limitation of time. Knowing the exact HDT may be more than we need. Knowing that we have plasticity or a particular E value at a given temperature may be eough to assume full cure. Hence I looked for some simplified notion.

    Our existing resin samples are round 30mm diameter x 5mm thick and not exact with the thickness. The inexact thickness is inconvenient, so is the higher Ixx of our sample. Our deflection at 0.2% strain as per the formal test is a lot less than their 0.25mm. Also with the wider section our warming rate may be different. So using the existing coupons to do some version of the HDT test may not be easy.

    Cheers
    Gregg.
     
  4. AndrewK
    Joined: Mar 2007
    Posts: 474
    Likes: 32, Points: 28, Legacy Rep: 344
    Location: Australia

    AndrewK Senior Member

    Viz
    I am also using R246 resin but the hardeners have different numbers here 126, 128, 160 etc. Sounds like you have done the same as I have formed your samples in plastic bottle tops, the milk tops here are nice and flexible with an inner rim seal so the discs pop out nice and easy.
    I found it best to keep the samples thin 2-3mm and as Jimbo mentioned grind down the meniscus edge. Drill a small hole close to the edge and hook a piece of wire through this and the other end bend over a cooking pot rim so the samples stay immersed in the water inside the pot. To speed things up you can preheat the water to about 10'C below the expected HDT from there on heat up the water slowly 2-3'C/min stirring with a thermometer. Quickly pull out a sample and test for softening by simply bending, this part will be subjective but I have found that the temperature spread from first sign of softening to very flexible is about 5'C. Middpoint correlates very well with the published data.
    If your HDT is above 100'C then you will need oil in the pot, cooking oil will do.

    Andrew
     
  5. viz
    Joined: May 2009
    Posts: 18
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Auckland NZ

    viz Junior Member

    Thanks Andrew,

    It's reasuring that you are just using the apparent softening/plasticity of the coupon as your test. I,m guessing this is the most common way. I can check with our resin supplier Adheasive Technologies here in AKL now the long weekend is over.
    Yes I too used milk bottle tops. For my first quick experiment I used an electric jug with the modern sealed element, a thermostat imersed and a thermometer next to that. If we can still easily use the thermostat I might try a pot on the stove like you and see if we can reduce the range of the temperature cycles.

    Now going out on a limb. It would be convenient to have a target min temp for apparent plasticity using these simplified methods (just bending coupon) as an indication of cure state. So we could ignore the warrming rate isue. The water would be held at target temp, the samples imersed for enough time to reach that temperature and then the bending test. The motive for going to more simplicity is mostly the large number of samples we take.

    Cheers
    Gregg.
     
  6. Jimbo1490
    Joined: Jun 2005
    Posts: 785
    Likes: 41, Points: 28, Legacy Rep: 527
    Location: Orlando, FL

    Jimbo1490 Senior Member

    Quick search on ebay found this controller, new in box with literature, and this RTD, also new. Even with shipping down to you guys, I bet you could still not break $100 US.

    Jimbo
     
  7. viz
    Joined: May 2009
    Posts: 18
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Auckland NZ

    viz Junior Member

    Thanks Jimbo,
    You didn't include the link to the controller you were refering to. I haven't had a look myself yet.
    We had a good but old controller (origionally given to us) which I overloaded last weekend. We have a spare internals for that but haven't got it to work yet. Maybe we will just replace it.

    Cheers
    Gregg
     

  8. AndrewK
    Joined: Mar 2007
    Posts: 474
    Likes: 32, Points: 28, Legacy Rep: 344
    Location: Australia

    AndrewK Senior Member

    Gregg,

    I take it that these are old samples that were heat treated alongside the job and you only need a pas fail test. Then as you suggest bringing the samples to a minimum temperature you are happy with and then checking to see if they are plastic or rigid is the simplest way of doing this.
     
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.