Видео

Thanks!

Thanks!

Glad it was useful!

Ha, I did the exact same but with a wire from a wire brush @ 0.5m Came out very even at 9% open area, with extremely small StDev

I do a lot of adhesion consulting across a broad range of adhesion problems. Feel free to contact me via my email (it's on my website) if you'd like to discuss.

@@stevenabbott3685 Wow！ great 👍, I will reach you soon from your website. Thanks for the reply!

Thanks for the kind feedback!

Thanks for the kind feedback!

I, and others, use it to explore "what if's". For example, I wanted to know if complicated multi-step roasts can be accomplished by 2-step roasts - which I why I added the 2-step output. I also think it helps explode a number of limitations with current roasting setups. So, rather than wave one's hands to talk about roasting, the model lets you see what is and isn't important in a roast. I've learned a lot from it and have realised that "first crack" is a non-event that should be called "last gasp". I wrote an article on this for Barista Hustle and, not surprisingly, it caused some fuss. First crack is supposed to be fundamental, yet the evidence is overwhelming that it's incidental. How to replace it? With a CO2 sensor that signals the onset of "chemistry".

@@stevenabbott3685 But isn't the CO2 release connected with the first crack? It is very interesting not to connect a very delicate procedure with an event (FC) that it is debatable and not easily distinguishable but then again, development time is still mallard reaction. Αny link for the article at Barista Hustle?

​@@georgevgenis1494 Plenty of roast don't crack, maybe only 10% of beans crack anyway, and there is no reason for CO2 onset and first crack to be strongly related. So let's switch to the significant signal, when significant chemistry is happening to the beans rather than an indirect signal that is only used because it's simple ... and because people assumed that the crack was significant whereas it's a last gasp.

@@stevenabbott3685 Couldn't agree more. That's why you see most roasters smelling coffee vigorously on either side of the first crack to smell the chemical reaction rather merely waiting for the FC and other physical observations.

@@GBRoasters Thanks for that. It's funny that in all the papers I've read, roasting videos I've watched and all the conversations I've had, no one has mentioned smelling the state of the roast. And that's a criticism of myself. It didn't occur to me to step outside my science box (focussing on CO2/CO chemistry) and wonder whether some sort of "aroma sensor" (other than the skilled human) could detect significant moments in the roast. Thanks for getting me out of my own bad thought habits! Steven

The Scheutjens-Fleer theory has been around since the 1970s so there's an extensive academic literature behind it. There's no single paper that captures all the above. Steven

Thank you for your kind comment - it encourages me to continue to help as best I can. Steven

Very much so. I created that video a long time ago. If you want an up-to-date guide then my Sticking Together book is affordable via Amazon and covers a wide range of issues, including core-shell rubber spheres in epoxies.

It's here: www.stevenabbott.co.uk/practical-surfactants/cmc.php. Good luck!

If you have the luxury of doing real calculations with, say, Comsol then you don't need these simple apps. As I say many times, "Adhesion is a property of the system" so knowing the system is the first requirement. In my simple apps I merely choose systems that have tractable formulae because I'm not smart enough to write anything that involves Finite Element stuff. Additive manufacturing is often hugely disappointing because interlayer adhesion can be catastrophic ... often because there's insufficient time/temperature to get polymer entanglement across the boundary. Certainly fused filament objects are often hopeless: temperatures high enough to guarantee entanglement make the whole part sag. Go to a lower temperature and the part stays the right shape, but entanglement is insufficient - I had to read a bunch of papers on the topic. So I don't think my apps help you directly, but if they help emphasise "system" thinking then they'll have achieved something!

If the weld is perfect then by definition we have a single piece of material and then "adhesion" is just the yield strength of the material - though I've never needed to know the fracture mechanics of how concentrated stresses get along the weld line. And if the weld is imperfect, say with a Griffith crack, then it's all about crack propagation. All way outside anything I know. Sorry!

I don't know the answer! I suppose for starters, most of don't have a convenient cryomill. But I think that you're right about "h". When you have a coating of a few μm on a larger particle, you can work with a reliable "h". But if you have a particle of uniform 10μm, we somehow have to change the equation to take into account spherical diffusion. And if there's a broad particle size distribution, that gets harder. But it's only "harder" not impossible. Maybe you can crack the problem and, as I said in the video, change the world of diffusion measurement forever!

They are a (hopefully) accurate measure of what the test provides, but they are otherwise highly misleading. And if they quote numbers such as X N/m2 then they are wrong because although the N are correct, the m2 are wrong because only a small part of the joint has experienced the stresses. In the book I talk of a few attempts to work out the "real" adhesion of the same joint measured using different techniques by experts ... and they simply can't be compared. A double lap joint does a better job at measuring shear failure adhesion, but it's a lot of work and it still tells you surprisingly little about what's going on - because Adhesion is a Property of the System. It's always better to do a test that relates to the specific failure modes experienced in your real application. Such tests won't be ASTM, but they will be much more useful.

The app is at www.stevenabbott.co.uk/practical-adhesion/wlf.php with an improved one at www.stevenabbott.co.uk/practical-adhesion/WLF-K.php. Although you can automatically fit C1, C2 and Tg, the point of the app (especially the newer one) is that you play with the parameteres to get a good-looking single curve - or, sometimes, the best-looking curve even if it's not super smooth. This is to help gain an understanding of what it all means.

@@stevenabbott3685 thank you verry much . Do you know how can i get master curve if i have creep conpliance data ..

Yes.By "combine" my priority would be to calibrate the app against a real Artisan chart then to validate it by changing conditions in the app and seeing what happens with the data from Artisan. But there are some complications. First, the app has its flaws which are steadily being fixed. Second, as I mention, the values measured by Artisan don't seem to me to be "real". Those with fancy external air supply roasters have a real IT so the app should work pretty well. But the majority have measured parameters that make little sense to me. The most recent version of the app has the TBT. I'd hidden this in previous versions because I didn't have confidence in it, but some academic reasearch suggests that it's correct. We all have a lot to learn, especially me!

The ultimate ​benefit of any roasting software is to make better tasting coffee. But taste preferences are different and roaster probes and measurements are as well. So from a sensory aspect the main question is, what to change in a roast profile to get wanted or to get rid of unwanted aromas. Yes, I think this would be very interesting: to validate it by changing conditions in the app and seeing what happens with the data from Artisan. And for a real workout the question remains whether the roasting machine is able to do the changes.

Sorry to reply in English. Scores are always subjective. That's not a weakness but a strength - you judge the sphere according to your own criteria, so the results are meaningful for your own needs. Steven

I know nothing of cocoa roasting. I will look it up! The basics of the app will apply to any drum roaster. The details such as water loss or special events equivalent to coffee first crack will be different. It would not be hard to generate a cocoa-specific app, but I need to read the literature first! Steven

I at last found time to check up on this. If you to chocolatealchemy.com/cocoa-bean-roasting you'll see that they use a coffee drum roaster AND have the same sort of BT curve and you roast to just beyond first crack. The BIG difference is that temperatures are MUCH lower, something like a 150C starting T rather than the 240C that's the default in the app. To allow a bigger range on the slider to set that T I then lose precision for coffee roasting. So I may have to make a separate cocoa version and change the settings and rewrite the next. Thanks for introducing me to another world I'd known nothing about!

Sorry it can only be screen captured

HLD is used around the world for many different types of formulation challenge and I use it myself regularly for problem solving. If you read my Surfactant Science free eBook you'll see that HLD is, of course, over-simplified and has its own problems, such as polar oils, but it beats all the non-alternatives every time!

Interesting point - and I don't know the answer. However, I can speculate that a thicker adhesive has a chance to fail cohesively, which is not a mode covered by the app. Because "adhesion is a property of the system" we have to think carefully about the overall effect of any one variable such as thickness, while here I'm only looking at stress concentrations, not the same thing as "adhesion".

Hi, did you find it?

In my (free) Particle Science book you'll see that neither D nor O thought that DLVO was a great theory. But I have often had to use this app as a starting point for deeper understanding of various particle phenomena. Steven

If you look up the Hansen Solubility Parameters for Chloroform they're 17.8 3.1 5.7. If we assume that this is a good match to your ABS, then solvents with similar values will be good. Interesting, both anisole and limonene are close, with anisole being a surprisingly green/safe solvent. If you wanted a more volatile solvent then cyclopentyl methyl ether is not so bad. You can find some more HSP values and some calculation capabilities from my site: www.stevenabbott.co.uk/practical-solubility/solvent-blends.php Good luck! Steven