On a sunny day (Wed, 20 Sep 2017 07:31:23 -0700) it happened John Larkin <jjlarkin@highlandtechnology.com> wrote in <rsu4scl6afrgq1pdc047f5ovo814uban0f@4ax.com>:>On Wed, 20 Sep 2017 07:48:29 GMT, Jan Panteltje ><pNaOnStPeAlMtje@yahoo.com> wrote: > >>On a sunny day (Tue, 19 Sep 2017 12:48:27 -0700) it happened John Larkin >><jjlarkin@highland_snip_technology.com> wrote in >><tts2scdlr3bgd9tqnajtv26d5n01rg4247@4ax.com>: >> >>>>Yes that is nice stuff, but those cemented ones as I use are >>>>very sturdy, although indeed in old TV sets (tube age) where those >>>>were used in series with heaters or power supply for example, >>>>they were a common fault source. >>> >>>Yes, they work a lot better with steady power dissipation. Spikes of >>>heat break them. The app where our big ones failed, they were acting >>>as sort of a pre-regulator to a big power supply, bang-bang mode. >>> >>>>At least they are explosion protected (by the alu casing). >>>>When your PCB thing explodes the parts will fly all over the place? >>>>Just to be nit-picking.. >>> >>>It's only 80 joules so I expect superficial wounds. >> >>What had been bothering me is minuscule pieces of metal flying about and shorting other critical electronics, >>perhaps control circuits or some overload protection. >>I have many modules from ebay that when inspected with a magnifying glass, >>show minuscule pieces of solder stuck to the board, >>when that comes lose and starts traveling, then no telling what can happen. >>Since that day I inspect every board with a magnifying glass. >> >>Jan Panteltje Low land Technology >>pick-a-second timing > >It's fun and pays the bills.It is truely 'low land' here, and the French call it Pays Bas, last week I was driving to a city next to here and noticed the fields had water, and the irrigation canals next to it were nearly overflowing, water on the road too. The county fired up the old steam powered pumps to get the water level down.. rain rain rain. And I am writing soft with .2 s resolution for my drone. No pico stuff. But,. the 'regulators' declared about 50 % of the country of limit for drones.... It is not fair. We need a revolution, this hinders inventivity and progress.
exploding wirewound resistors
Started by ●September 15, 2017
Reply by ●September 20, 20172017-09-20
Reply by ●September 20, 20172017-09-20
On Wed, 20 Sep 2017 06:37:09 -0400, Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:>On 09/19/2017 05:55 PM, John Larkin wrote: >> On Tue, 19 Sep 2017 11:02:38 -0700, John Larkin >> <jjlarkin@highland_snip_technology.com> wrote: >> >>> On Fri, 15 Sep 2017 13:16:39 -0700, John Larkin >>> <jjlarkin@highland_snip_technology.com> wrote: >>> >>>> >>>> >>>> Lots of people make parts like this: >>>> >>>> https://www.digikey.com/product-detail/en/bourns-inc/PWR5322WR250JE/PWR5322WR250JETR-ND/2022962 >>>> >>>> The overload specs are almost always 5x rated power for 5 seconds, so >>>> a 3 watt part can absorb 15 joules in that time. I wonder what happens >>>> for shorter time overloads? Constant joules? >>>> >>>> Has anyone experimented with stressing this sort of small wirewound >>>> resistor? >>>> >>>> I'll probably order some and blow them up. We'll be using low ohms, >>>> 0.2 maybe, so we'll need a lot of current. There's a car repair place >>>> and a motorcycle shop on our block, so maybe I could buy or borrow a >>>> battery. For pure fast joules, I could just discharge some big >>>> electrolytics. >>> >>> OK, I'm ready to blow up some resistors, as soon as UPS arrives. >>> >>> https://www.dropbox.com/s/u2078h7upn81q10/Z420_3.JPG?raw=1 >>> >>> That should do numbers like 300 amps, 80 joules, 50 volts. >> >> We got some TE 3-watt parts, 0.33R, SMW3 series. I kept increasing >> joules per shot. At 10.8J, the resistor went POP but kept working. The >> case has a small crack on top and little puffs of smoke came out every >> pulse. >> >> It seems to have run out of smoke now. Still works. > >There's also non-magic smoke. > >Cheers > >Phil HobbsI'm conjecturing that the nichrome element is now nestled in a little coccoon of carbon, not much left to boil out. I need a MELF wirewound, exposed wire on the outside and ceramic inside. -- John Larkin Highland Technology, Inc lunatic fringe electronics
Reply by ●September 20, 20172017-09-20
Den onsdag den 20. september 2017 kl. 17.23.27 UTC+2 skrev John Larkin:> On Wed, 20 Sep 2017 06:37:09 -0400, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > > >On 09/19/2017 05:55 PM, John Larkin wrote: > >> On Tue, 19 Sep 2017 11:02:38 -0700, John Larkin > >> <jjlarkin@highland_snip_technology.com> wrote: > >> > >>> On Fri, 15 Sep 2017 13:16:39 -0700, John Larkin > >>> <jjlarkin@highland_snip_technology.com> wrote: > >>> > >>>> > >>>> > >>>> Lots of people make parts like this: > >>>> > >>>> https://www.digikey.com/product-detail/en/bourns-inc/PWR5322WR250JE/PWR5322WR250JETR-ND/2022962 > >>>> > >>>> The overload specs are almost always 5x rated power for 5 seconds, so > >>>> a 3 watt part can absorb 15 joules in that time. I wonder what happens > >>>> for shorter time overloads? Constant joules? > >>>> > >>>> Has anyone experimented with stressing this sort of small wirewound > >>>> resistor? > >>>> > >>>> I'll probably order some and blow them up. We'll be using low ohms, > >>>> 0.2 maybe, so we'll need a lot of current. There's a car repair place > >>>> and a motorcycle shop on our block, so maybe I could buy or borrow a > >>>> battery. For pure fast joules, I could just discharge some big > >>>> electrolytics. > >>> > >>> OK, I'm ready to blow up some resistors, as soon as UPS arrives. > >>> > >>> https://www.dropbox.com/s/u2078h7upn81q10/Z420_3.JPG?raw=1 > >>> > >>> That should do numbers like 300 amps, 80 joules, 50 volts. > >> > >> We got some TE 3-watt parts, 0.33R, SMW3 series. I kept increasing > >> joules per shot. At 10.8J, the resistor went POP but kept working. The > >> case has a small crack on top and little puffs of smoke came out every > >> pulse. > >> > >> It seems to have run out of smoke now. Still works. > > > >There's also non-magic smoke. > > > >Cheers > > > >Phil Hobbs > > I'm conjecturing that the nichrome element is now nestled in a little > coccoon of carbon, not much left to boil out. > > I need a MELF wirewound, exposed wire on the outside and ceramic > inside. >http://www.firstohm.com.tw/phocadownloadpap/01_resistors/SWM20160721.pdf
Reply by ●September 20, 20172017-09-20
On Tuesday, September 19, 2017 at 10:53:49 PM UTC-4, John Larkin wrote:> > I worked on a kilojoule flash once. It was a couple hundred pounds of > oil caps and power supply on a cart. It was used to develop Kalvar > film, which is (or was) diazo stuff embedded in plastic. A big light > flash makes microbubbles, which scatters light. > > Geez, I have done so much weird stuff. > > > -- > > John Larkin Highland Technology, Inc > > lunatic fringe electronicsAh yes. And I have invested in weird things. Owned a couple of shares of Kalvar at one time. Oh well that is how one learns. Dan
Reply by ●September 20, 20172017-09-20
John Larkin wrote:> Really, in the last couple of weeks, we've explored a zillion options, > including fuses, self-protecting fets, current limiting the fets, all > sorts of crazy schemes... each with at least one fatal flaw.A tiny magamp, perhaps? The cover story is my high current synchronous rectifier based on this principle, the photo explains the main idea of FET current measurement. https://www.google.pl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0ahUKEwjMu-Sc1rTWAhWlNJoKHWeIC4kQFggvMAE&url=http%3A%2F%2Fm.publio.pl%2Ffiles%2Fsamples%2F84%2Fca%2F83%2F154876%2FElektronika_dla_Wszystkich_1_demo.pdf&usg=AFQjCNFO-ibNX2OB6AcZjp7LQS184-JkYQ Best regards, Piotr
Reply by ●September 20, 20172017-09-20
On Wed, 20 Sep 2017 23:12:22 +0200, Piotr Wyderski <peter.pan@neverland.mil> wrote:>John Larkin wrote: > >> Really, in the last couple of weeks, we've explored a zillion options, >> including fuses, self-protecting fets, current limiting the fets, all >> sorts of crazy schemes... each with at least one fatal flaw. > >A tiny magamp, perhaps? The cover story is my high current >synchronous rectifier based on this principle, the photo >explains the main idea of FET current measurement. > > >https://www.google.pl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0ahUKEwjMu-Sc1rTWAhWlNJoKHWeIC4kQFggvMAE&url=http%3A%2F%2Fm.publio.pl%2Ffiles%2Fsamples%2F84%2Fca%2F83%2F154876%2FElektronika_dla_Wszystkich_1_demo.pdf&usg=AFQjCNFO-ibNX2OB6AcZjp7LQS184-JkYQ > > Best regards, PiotrWish I could read that! Got a schematic? I could probably understand that. The tiny toroids around the fet source leads are cute. We are sensing the voltage across the fets plus a series resistor. We added a resistor because a passive resistor can absorb a lot more joules than a mosfet, especially for short times, until we can shut down the gate drive. -- John Larkin Highland Technology, Inc lunatic fringe electronics
Reply by ●September 21, 20172017-09-21
On Wed, 20 Sep 2017 08:57:59 -0700 (PDT), Lasse Langwadt Christensen <langwadt@fonz.dk> wrote:>Den onsdag den 20. september 2017 kl. 17.23.27 UTC+2 skrev John Larkin: >> On Wed, 20 Sep 2017 06:37:09 -0400, Phil Hobbs >> <pcdhSpamMeSenseless@electrooptical.net> wrote: >> >> >On 09/19/2017 05:55 PM, John Larkin wrote: >> >> On Tue, 19 Sep 2017 11:02:38 -0700, John Larkin >> >> <jjlarkin@highland_snip_technology.com> wrote: >> >> >> >>> On Fri, 15 Sep 2017 13:16:39 -0700, John Larkin >> >>> <jjlarkin@highland_snip_technology.com> wrote: >> >>> >> >>>> >> >>>> >> >>>> Lots of people make parts like this: >> >>>> >> >>>> https://www.digikey.com/product-detail/en/bourns-inc/PWR5322WR250JE/PWR5322WR250JETR-ND/2022962 >> >>>> >> >>>> The overload specs are almost always 5x rated power for 5 seconds, so >> >>>> a 3 watt part can absorb 15 joules in that time. I wonder what happens >> >>>> for shorter time overloads? Constant joules? >> >>>> >> >>>> Has anyone experimented with stressing this sort of small wirewound >> >>>> resistor? >> >>>> >> >>>> I'll probably order some and blow them up. We'll be using low ohms, >> >>>> 0.2 maybe, so we'll need a lot of current. There's a car repair place >> >>>> and a motorcycle shop on our block, so maybe I could buy or borrow a >> >>>> battery. For pure fast joules, I could just discharge some big >> >>>> electrolytics. >> >>> >> >>> OK, I'm ready to blow up some resistors, as soon as UPS arrives. >> >>> >> >>> https://www.dropbox.com/s/u2078h7upn81q10/Z420_3.JPG?raw=1 >> >>> >> >>> That should do numbers like 300 amps, 80 joules, 50 volts. >> >> >> >> We got some TE 3-watt parts, 0.33R, SMW3 series. I kept increasing >> >> joules per shot. At 10.8J, the resistor went POP but kept working. The >> >> case has a small crack on top and little puffs of smoke came out every >> >> pulse. >> >> >> >> It seems to have run out of smoke now. Still works. >> > >> >There's also non-magic smoke. >> > >> >Cheers >> > >> >Phil Hobbs >> >> I'm conjecturing that the nichrome element is now nestled in a little >> coccoon of carbon, not much left to boil out. >> >> I need a MELF wirewound, exposed wire on the outside and ceramic >> inside. >> > >http://www.firstohm.com.tw/phocadownloadpap/01_resistors/SWM20160721.pdfI emailed them about joule rating, or samples to test. The coating looks organic. Maybe they would sell me uncoated ones. -- John Larkin Highland Technology, Inc lunatic fringe electronics
Reply by ●September 21, 20172017-09-21
On Thursday, 21 September 2017 04:16:01 UTC+1, John Larkin wrote:> On Wed, 20 Sep 2017 08:57:59 -0700 (PDT), Lasse Langwadt Christensen > <langwadt@fonz.dk> wrote: > >Den onsdag den 20. september 2017 kl. 17.23.27 UTC+2 skrev John Larkin: > >> On Wed, 20 Sep 2017 06:37:09 -0400, Phil Hobbs > >> <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> >On 09/19/2017 05:55 PM, John Larkin wrote:> >> >> We got some TE 3-watt parts, 0.33R, SMW3 series. I kept increasing > >> >> joules per shot. At 10.8J, the resistor went POP but kept working. The > >> >> case has a small crack on top and little puffs of smoke came out every > >> >> pulse. > >> >> > >> >> It seems to have run out of smoke now. Still works. > >> > > >> >There's also non-magic smoke. > >> > > >> >Cheers > >> > > >> >Phil Hobbs > >> > >> I'm conjecturing that the nichrome element is now nestled in a little > >> coccoon of carbon, not much left to boil out. > >> > >> I need a MELF wirewound, exposed wire on the outside and ceramic > >> inside. > >> > > > >http://www.firstohm.com.tw/phocadownloadpap/01_resistors/SWM20160721.pdf > > I emailed them about joule rating, or samples to test. > > The coating looks organic. Maybe they would sell me uncoated ones.Then you'd get damp ingress, corrosion etc. You need to go back to the 1920s for uncoated resistors. NT
Reply by ●September 21, 20172017-09-21
John Larkin wrote:> Wish I could read that! Got a schematic? I could probably understand > that. The tiny toroids around the fet source leads are cute.https://s26.postimg.org/lk8iapmnd/prostownik08.png "Obciazenie" is for "Load". It works as follows: the IR21531 is just a pump feeding the inductive elements with 12V/~400kHz. The saturable reactors S1-S4 form LR voltage dividers with their respective resistors, the not shown MOSFET gate capacitances of ~27nF each act as the filter caps for the rectifiers feeding the gates. The gates are kept low by the resistors. All the FETs are initially disabled. When the load current flowing through the MOSFETs' body diodes exceeds a certain value, the reactance of the S reactors drops as it approaches the core saturation region, raising the voltage at the gate and eventually makes the MOSFET conduct, absorbing the diode current. So the more it is loaded, the more synchronous it is. At ~1.5A the MOSFET pairs are fully conducting. The low Vth T5 and T6 are just a crossconduction protection circuitry and are passive during normal operation. This is just a very stable 50Hz bridge rectifier with fully analog control and heavy (ab)use of the otherwise unwanted phenomena: huge gate charge, body diode, magnetic saturation region etc. I've been using it for about 2 years 24/365 to power a bunch of low voltage halogen lamps at about 50A. Best regards, Piotr
Reply by ●September 21, 20172017-09-21
Piotr Wyderski wrote:> The saturable reactors S1-S4 form LR voltage dividers with their > respective resistorsThis is what happens, but the description is a bit vague. The voltage across the Tr2 and Tr3 primaries depends on the flux in S1-S4, which in turn depends on the magnetizing current flowing through the source leads (the leads form one-turn control windings of the magamps). So the final control mechanism depends on both LL (e.g. S1+S2+Tr3) and LR divider actions. Best regards, Piotr
