You're welcome. I watched this phenomena when I designed the inductor for a capacitive Discharge Ignition used in Honda Engine conversions for Homebuilt Aircraft. It was pretty easy to create a design that had 85mJ of energy and 350V stored into a capacitor. What was more difficult was getting specifications on inductor core saturation. I started with small inductor cores that became larger and larger until finally I ended up using two large toroids stacked with a proper number of turns. I then was able to run 10A into this inductor before it started saturating. That energy, when the current was shut off, was dumped into the capacitor and then when the time was right, the 350V was applied through a triac into the ignition coil. One awesome spark with a battery voltage of 8 volts during cranking at -40C. Regards, John Dammeyer Wireless CAN with the CANRF module now available. http://www.autoartisans.com/products Automation Artisans Inc. Ph. 1 250 544 4950 > -----Original Message----- > From: pic microcontroller discussion list > [mailto:PICLIST@MITVMA.MIT.EDU] On Behalf Of Jai Dhar > Sent: Wednesday, March 05, 2003 10:09 AM > To: PICLIST@MITVMA.MIT.EDU > Subject: Re: [EE]: Project testing > > > Thank you John, that was exactly what I was looking for... a > reasonable > explanation as to why my regulator would blow out. I knew all > the theoretical > characteristics of inductors, but unfortunately have not been > exposed to the > practical side yet. Veroboard will have to do since I don't > have access to > board making accessories, however, I will definitely replace > that inductor > with one capable of handling an appropriate amount of > current. This should > hopefully help. > > Thank you all, > > Jai > > Quoting John Dammeyer : > > > Hi Jai, > > > > Switching supplies can be incredibly simple with the new > devices that > > are out there and they can easily deliver the expected > current with the > > suggested circuit from the Maxim data sheet. > > > > However, there are some fundamental rules that need to be followed. > > You're dealing with a high frequency, high power design and > so circuit > > layout becomes an important issue. That means if you are having > > problems, the breadboard or vero board would be the first suspicious > > element to remove. > > > > *** So a proper layout is very important to how well this type of > > circuit will function. *** > > > > I haven't checked the recommended inductor on the data > sheet but usually > > a number of inductor manufacturers and their part numbers > are listed. > > If sounds like, from your response on the inductor issue > that you aren't > > clear on saturation current of an inductor. I suspect you > don't have a > > scope to look at waveforms either. > > > > So a bit of theory. When you apply a voltage across an inductor and > > measure the voltage at that instant in time, you will measure the > > applied voltage. If at that same instant in time you measure the > > current it will be 0. After an infinite amount of time the > voltage will > > be zero and the current will be infinite. That's the theoretical > > inductor. > > > > In reality, the inductor has a certain amount of resistance and the > > power supply has limitations on current so one of two things will > > happen: the current increases to a point where the I^2R of > the inductor > > wire is exceeded and it heats up and goes open circuit; the power > > supply goes into current limit and holds the current at Xi and the > > voltage drops to the I*R value across the inductor. > > > > So far pretty simple. > > > > What happens in-between is what is interesting. Simply > put, if you were > > to watch the current flow on a scope you'd see an > exponential increase > > in current but when the magnetic field created by this > current saturates > > the core, there will be a dramatic almost avalanche like > increase in > > current without any subsequent increase in energy storage. The coil > > heats up and could toast the driver transistor. > > > > So if you are using a 47uH inductor that saturates at 20ma > then trying > > to run 500ma through the inductor will probably toast the regulator > > because the current will increase way beyond the > specifications of the > > device before it has a chance to shut down the current; unless the > > inductor burns out first. > > > > When specifying an inductor, make sure that the current > rating of the > > inductor is at least the value of the regulation device. > For example: > > The Talema inductors listed as being suitable for National > Semi Simple > > Switchers in the DigiKey catalog has and inductor 54uH at > 1A and another > > one 53 uH at 3A. The amperage rating has far more to do with core > > saturation than wire size and heating since the 1A coil has a DCR of > > 0.20 Ohms while the 3A device has a DCR of 0.13 Ohms. > > > > Hope that helps. > > > > John Dammeyer > > > > > > Wireless CAN with the CANRF module now available. > > http://www.autoartisans.com/products > > Automation Artisans Inc. > > Ph. 1 250 544 4950 > > > > > > > -----Original Message----- > > > From: pic microcontroller discussion list > > > [mailto:PICLIST@MITVMA.MIT.EDU] On Behalf Of Jai Dhar > > > Sent: Wednesday, March 05, 2003 8:42 AM > > > To: PICLIST@MITVMA.MIT.EDU > > > Subject: Re: [EE]: Project testing > > > > > > > > > Quoting "Alan B. Pearce" : > > > > > > > >As for the comment made about breadboards being the > worst way to > > > > >go... keep in mind I'm just testing it now, I don't > plan on boxing > > > > >this up and using it with a breadboard :-) > > > > > > > > I suspect the problem is the breadboard. This is allowing > > > all sorts of > > > > horrible lead lengths at places they should not be, which > > > is allowing the > > > > switchmode chip to oscillate in all sorts of uncontrolled > > > ways, and this is > > > > what has destroyed your chips. > > > > > > > > The only way you are going to sort this is to go straight > > > to a PCB which has > > > > been laid out with current paths and suitable sized tracks > > > straight off. > > > > Then you will get something that will approach the > > > stability you require. > > > > > > > > > > Does veroboard have tracks suitable for this application? > > > Seems to be my only > > > option now, unless I reduce the capabilities of my supply. > > > > > > As to the type of inductor that I'm using, what is the > > > 'correct' type? I > > > wasn't aware that a special type of inductor was required for this > > > application, so I would appreciate knowing what type of > > > inductor I ned to > > > obtain. > > > > > > Regarding Jochen mentioning that I need a bigger > capacitor after the > > > rectifier... do I really need something as big as 4700? The > > > biggest cap I have > > > seen used after a rectifier is a 2200 (at least looking > at the example > > > schematics on psu's). And yes, I did use heatsinks :-) > > > > > > While all these are great suggestions in improving my design, > > > are any of these > > > explanations as to why my regulator fried when I simply put > > > in an appropriate > > > resistor? It seems like all these suggestions could *improve* > > > my design, but > > > from what I understand, none of them seem to address why > > > drawing 500mA would > > > fry my regulator. I must be missing something. > > > > > > Nevertheless, thank you for all of your input thus far, > > > > > > Jai > > > > -- > > > > http://www.piclist.com hint: The PICList is archived > three different > > > > ways. See http://www.piclist.com/#archives for details. > > > > > > > > > > > > > > > > > > > ---------------------------------------- > > > This mail sent through www.mywaterloo.ca > > > > > > -- > > > http://www.piclist.com hint: The PICList is archived > three different > > > ways. See http://www.piclist.com/#archives for details. > > > > > > > > > > -- > > http://www.piclist.com hint: The PICList is archived three different > > ways. See http://www.piclist.com/#archives for details. > > > > > > > ---------------------------------------- > This mail sent through www.mywaterloo.ca > > -- > http://www.piclist.com hint: The PICList is archived three different > ways. See http://www.piclist.com/#archives for details. > > -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.