answers to those doggone thermal design questions
by tony kordyban
dear doggone thermal guy,
our company is a giant manufacturer of specialty sportswear fabrics and lenses for automotive taillights (the ones that smash so easily when a grocery cart taps them in the parking lot). we recently entered the electronics cooling business when our ceo saw a story in the wall street journal about microprocessor power doubling every three to five years. we think our carbon fiber impregnated plastics, which didn't quite take off as anti-sharkbite armor or black-shaded taillight lenses, would make great heat transfer interfaces for electronic assemblies.
my job is to get the word out. i have been manning booths at trade shows and putting our name on magazine bingo cards for months. the three engineers i finally talked to were thrilled with the possibilities of our products, but it's harder to get our information into the hands of a thermal engineer than to feed brussels sprouts to a four-year old.
how do you look for thermal product information when you need it? if i knew that, i'd know where to place it.
joe gladhand from plaidsville
i think i saw your booth last summer. was that the one that gave away the glow-in-the-dark cardboard slide rule for converting f to c degrees? i think it's at the bottom of my junk drawer in the lab.
it used to be i'd keep a whole file cabinet of suppliers' catalogs, and a pencil drawer full of sales rep business cards. it used to pay to know the biggest pack rat in the office, the geek who actually kept all his catalogs in alphabetical order.
but then they invented the internet, and more specifically, the world wide web. after i learned how to use it, and then, after many of my suppliers learned how to use it as well, all those catalogs, brochures, industrial directories and business cards went straight into the recycling bin.
now, when i need data on heat sinks, fans, thermal interface materials, or electronic component packaging, i go straight to the vendor's web site. no calling a local sales rep, no playing phone tag with an application engineer, no begging a literature distribution clerk to send me a catalog that is already three years out of date. a good vendor's web site has all its product data available to me at the click of a mouse. a poor web site has just a nice photo of the company headquarters and a list of phone numbers to call for more information.
notice i didn't say the vendors or their products were good or bad. some of the best products have the worst web sites, with absolutely no technical information on them. or worse yet, they require you to register and get a password just to look at their product data sheets. i hate to say it, but when i am searching for new product information, and i hit a web site that doesn't want to give me the information i want, i quickly jump to the next one on my list, no matter how good the vendor might be. i admit it. i am spoiled. give me my data now, or just forget about it. i don't want to wait for a return phone call or for somebody to fax me the datasheet. show me the datasheet on line.
so, joe, keep going to the trade shows, keep putting ads in magazines, and keep cold-calling prospects. all that still has a place. but in addition to that, make a good web site. put everything, and i mean everything that any customer might conceivably want to know about your products on it. forget the fancy animation and 3-d graphics -- make the darn web site searchable. maybe i want to know who some of your well known customers are. maybe i want to know if you offer a thermal grease that is transparent to infrared radiation. i should be able to find out either one of those things in less than 60 seconds.
such a website will not be easy to put together, so don't leave it to your kid in junior high. get some professionals to put it together. there are lots of former dot-commers out there looking for web design work. have real customers try it out, and if they complain, listen to them. once you have a decent web site, make sure we can find it. pay for a good web address. i think ibm uses www.ibm.com, for one obvious example. then pay the top search engines to list your site. i know they ought to do it for free, but if you want to be near the top of the list of 15,000 web sites that pop up when i search for the phrase "carbon fibers", you will have to shell out a few bucks.
that's my advice to joe. but i'll throw out a question to the other readers. am i nuts? not in general, but about this web site thing? do you find yourself doing more and more of your thermal product data shopping on the web, or do you still wait for printed catalogs and brochures to come through the mail? and do you have any particularly good or bad vendor web sites you'd like to share?
a couple of years ago when sales were great, the biggest problem my management had was deciding between domestic and imported champagne at the stock option parties. now management is looking to cut costs and everybody else is scrambling to justify their existence. my manager, like most, has a vague idea that keeping circuits from getting too hot is a good idea, but how can i convince him that thermal engineering has a direct benefit on the bottom line of the company?
scrambler from scranton
i hear you, friend. for too long the thermal engineer has been the most invisible member of the electronic hardware design team. unfortunately, when times get tough, the only visibility the thermal engineer has is in the payroll department. here is the e-mail i'd send to my manager arguing for not cutting the thermal engineer, or perhaps even looking to add more:
dear bottom line manager,
we are not in the business of selling thermal design solutions. we design and sell electronic hardware. our customers don't care if it has the latest optimum heat sink, they just want it to work, and for the right price. so it's easy for anybody to think that the work done by the thermal engineer doesn't contribute much to the bottom line.
maybe 100 watts on a single chip doesn't sound like a lot to you. i'll give you a sense of scale. at 0.1 watt on a chip, you don't need much of anything to keep it cool. that's what chips were like in the 70's and 80's when we didn't have a thermal engineer. at 1.0 watt per chip, you have to be a little careful how you lay out components on a board to keep them from getting too hot. at 10 watts per chip, you need somebody who knows what he is doing to select the right heat sink. at 100 watts per chip, you need a thermal expert involved with the design from concept all the way through manufacture to make sure your product will not be killed by thermal problems. above 100 watts per chip, (not far away on the current roadmap) you need a miracle.
so we are faced with a dilemma for new product development. if we don't use the latest technology chips that are being introduced, then our products will be slower, bulkier, and support fewer features than our competitor's products. in a short time we won't have any customers, and therefore no bottom line.
so i guess we will have to give in and use the latest available chips with their ever-increasing power. but without that thermal engineer on hand, those high power chips just aren't going to work. or we might have to spin our hardware a couple of extra times to work out the thermal problems by trial and error.that would probably make us late to market. and late means no customers again.
either way, no customers, no bottom line.
maybe we could get away with outsourcing the thermal work. all we'd have to do is make sure that high-priced consultant is there from day one to advise the board designers, the chassis designers, and whoever will put together our cooling system, so that all the little thermal issues can be ironed out as the design process goes along. if you just bring him in at the last minute, he will find all our thermal problems all right, and then he'd send us back to square one to re-design everything to fix them. the days of just slapping on a heat sink or a fan are long gone. thermal design has to be integrated into the hardware design process, or we don't have a hardware design process. it is actually cheaper and faster to have the thermal engineer in house, working with the other engineers on a daily basis. faster and cheaper than doing the whole thing over, that is.
faster to market. cheaper. isn't that the bottom line?
your heat-sink-pickin', fan-slappin' thermal engineer
if you had a chance to attend the coolingzone conference, cooling electronics -- the next decade, you would have heard this story over and over. if your company is going to have a future, it will have to make use of the electronic components of the future. and those components will require expert thermal design to keep them from burning up.
thermal engineers won't knock a mere 2% off the cost of an electronic assembly. they will provide an enabling technology for the electronics of the future.
about tony kordyban
tony kordyban has been an engineer in the field of electronics cooling for different telecom and power supply companies (who can keep track when they change names so frequently?) for the last 20 years. maybe that doesn't make him an expert in heat transfer theory, but it has certainly gained him a lot of experience in the ways not to cool electronics.
he does have some book-learnin', with a b.s. in mechanical engineering from the university of detroit and a master’s in mechanical engineering from stanford. in those 20 years tony has come to the conclusion that a lot of the common practices of electronics cooling are full of baloney. he has run into so much nonsense in the field that he has found it easier to just assume "everything you know is wrong" (from the comedy album by firesign theatre), and to question everything against the basic principles of heat transfer theory.
tony has been collecting case studies of the wrong way to cool electronics, using them to educate the cooling masses, applying humor as the sugar to help the medicine go down. these have been published recently by the asme press in a book called, "hot air rises and heat sinks: everything you know about cooling electronics is wrong." it is available at https://www.amazon.com/hot-air-rises-heat-sinks/dp/0791800741. this advice column is an extension of that educational effort.