300W Resistive load?
in [Design]
|
From: Grant on 18 Jul 2010 18:45 On Sun, 18 Jul 2010 20:27:27 GMT, nico(a)puntnl.niks (Nico Coesel) wrote: >"Tim Williams" <tmoranwms(a)charter.net> wrote: > >>"Nico Coesel" <nico(a)puntnl.niks> wrote in message = >>news:4c431fff.249648812(a)news.planet.nl... >>> If you use transistors that can handle 200V (or way more) like I did >>> there is no problem. >> >>I'm just wondering if they were MOSFET or bipolar. TO-3 MOSFETs are = >>somewhat old and rare. MJ15025 for instance won't handle more than = >>0.17A at 200V, according to the SOA, which is only 34W, out of a rated = >>250W. > >The ones I used are bipolar. I got them from a 150A or 300A 5V PSU. I >don't know the part number but IIRC it is some sort of special type >for switching purposes anyway. Each transistor is bolted to the >heatsink directly for optimum heat transfer (the heatsink is live!). > >My dummy load actually is an adjustable current sink. A potmeter, a >darlington emittor follower and series resistor make the drive for the >'end stage'. > >The maximum load I ever tested was 400W. But not up at 200V? Low voltage, high current is easy, transistor SOA at high voltage not so easy? I made a controlled load, eight transistors on one side of a two piece box type heatsink, eight 0R33, 50W resistors series/parallel on other side, with a fan at one end, could do up to 600W (30V @ 20A) loads. Live heatsink 'cos insulator cost too much in thermal resistance. Needed a heat spreader piece of 3mm thick aluminium between transistors to heatsink to help the heat transfer. Grant.
From: Jim Thompson on 18 Jul 2010 21:39 On Sun, 18 Jul 2010 17:17:09 -0500, John Fields <jfields(a)austininstruments.com> wrote: >On Sat, 17 Jul 2010 17:06:10 -0700, Jim Thompson ><To-Email-Use-The-Envelope-Icon(a)On-My-Web-Site.com> wrote: > >>On Sun, 18 Jul 2010 09:30:22 +1000, Grant <omg(a)grrr.id.au> wrote: >> >>>On Sat, 17 Jul 2010 09:25:42 -0700, Joerg <invalid(a)invalid.invalid> wrote: >>> >>>>Hammy wrote: >>>>> On Sat, 17 Jul 2010 07:37:28 -0700, Joerg <invalid(a)invalid.invalid> >>>>> wrote: >>>>> >>>>>> Hammy wrote: >>>>>>> On Sat, 17 Jul 2010 07:00:35 -0700, Joerg <invalid(a)invalid.invalid> >>>>>>> wrote: >>>>>>> >>>>>>>> Hammy wrote: >>>>>>>>> What is something cheap and simple that can be used to dissapate 300W >>>>>>>>> on a 200Vdc bus. I'm trying to think of something I could maybe pick >>>>>>>>> up at a local homedepot or similiar. >>>>>>>>> >>>>>>>>> I need a 300W load to test out my PFC. >>>>>>>> Light bulbs? Two in series and then as many of those in parallel as needed. >>>>>>> Sounds good one 60W bulb is about 200 ohms. ... >>>>>> >>>>>> At 200V each bulb gets only 100V so the resistance will be a little >>>>>> lower. You could measure that if it must be precise. >>>>> >>>>> I don't need high precision if I can test at even 80% rated load and >>>>> maybe do some pulse test and off course use a CT to check the input >>>>> current I'll be happy. >>>> >>>> >>>>Well, just wanted to mention that because when at undervoltage the bulbs >>>>will present a higher load. Also, when you start any kind of converter >>>>from zero they will almost look like a short initially. Same if you >>>>connect a cold bulb (cold filament) to a running converter. It's not >>>>like a power resistor. >>>> >>>> >>>>>>> ... I'll look around for some small cheap bulb mounts. >>>>>> >>>>>> You can also solder to them. If careful they can later still be used for >>>>>> their intended purpose. But make sure you place them onto a heat >>>>>> resistant surface during the test :-) >>>>> >>>>> Your taking all the fun out of this :-) >>>>> >>>> >>>>The fun might come back when connecting the cold bulbs not two by two, >>>>but all at once. Tchk ... *POOF*. >>> >>>Decent converter should handle that? Not like 8:1 lamp hot:cold is >>>a short circuit, almost a short ;) >>> >>> >>>I keep an old fuser element in the cupboard for hi voltage loads, and >>>have light bulbs setup for 24V with 12V halogens in series, then proper >>>high wattage resistors picked up over time. >>> >>>Lamps are easiest to hook up, perhaps three or more in series at first? >>> >>>Over here they've banned incandescent lamps already, only those new and >>>expensive hi-voltage halogens available :( >>> >>>What about an electric heater on low? Got a couple or more to wire >>>in series? Raid an electric stove, put elements in series? >>>> >>>> >>>>>> Good news is you may not even have to drive all the way to Home Depot. >>>>>> Supermarkets also have a variety of them. Of course, then your wife >>>>>> might hand you a lengthy "while you are there anyhow" list and you don't >>>>>> get to look at all the new power tools, barbecues and stuff. Man's toy >>>>>> store ... >>>>> We call that a honey do list here.;-) >>>>> >>>> >>>>Nah, a honey-do list isn't grocery shooping, it's when the missus wants >>>>you to get that tree stump out over yonder, and it's 105F :-) >>> >>>Grant. >> >>I've been known to take two sheets of copper clad, drill a grid of >>holes and solder a hundred or so 2W resistors between the planes, put >>a muffin fan on it and away you go... :-) > >Why copper clad (assuming you meant something like G-10 from back then >(FR-4 today)) when copper sheet would have been so much cheaper and, >for the same thickness, would have had _much_ better thermal >conductivity? > >JF Why copper sheet, when I had G-10 all over the place? ...Jim Thompson -- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 |
From: Uwe Hercksen on 19 Jul 2010 05:07 Hammy schrieb: > > Sounds good one 60W bulb is about 200 ohms.I'll look around for some > small cheap bulb mounts. Hello, the resistance of the bulb depends heavily on the filament temperature, take a resistance meter and measure the cold 60 W bulb, you will be surprised. Bye
From: Hammy on 19 Jul 2010 08:10 On Mon, 19 Jul 2010 11:07:01 +0200, Uwe Hercksen <hercksen(a)mew.uni-erlangen.de> wrote: > > >Hammy schrieb: >> >> Sounds good one 60W bulb is about 200 ohms.I'll look around for some >> small cheap bulb mounts. > >Hello, > >the resistance of the bulb depends heavily on the filament temperature, >take a resistance meter and measure the cold 60 W bulb, you will be >surprised. > >Bye Hi Uwe I already knew that, and it was also brought up several times in the thread. Their are a few reasons why I'm not concerned about it I'm useing a 500VA isolation transformer (winding resistance) and the measured resistance from the fuse to the bridge rectifier is in excess of 4 oHms this includes a 2.5 oHm NTC. So when you consider all the R the inrush current cannot be above 35 to 40A worst case. This is a boost topology so whats in the inrush path? the bridge and the PFC choke and the output rectifier. The rectifiers are rated for 100A or better surge current. The inductor saturates at start-up, so what it would anyway just from the inrush from the 390uF output cap. The controller wont start probably until after the surge anyways,if it were actually to start it should enter hiccup mode.
From: Tim Williams on 19 Jul 2010 09:40
"Grant" <omg(a)grrr.id.au> wrote in message news:g90746dqa83huenng5oce10se4lnssi6v7(a)4ax.com... > But not up at 200V? Low voltage, high current is easy, transistor > SOA at high voltage not so easy? > > I made a controlled load, eight transistors on one side of a two piece > box type heatsink, eight 0R33, 50W resistors series/parallel on other > side, with a fan at one end, could do up to 600W (30V @ 20A) loads. The nice thing about high voltage is, you can just cascode them. Stack collector to emitter and bases to voltage dividers. Put a rheostat in the bottom emitter to control current. Then you don't have to worry about emitter resistors at all, and 2nd breakdown is now hundreds of volts away. You get shitty hFE on most HV transistors, so the bottom transistors carry more current, but the base voltage divider will sag under load, approximating equal power dissipation for some loads. So it's not terrible. Tim -- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms |