On Tuesday 01 April 2008, Jon Elson wrote: >Gene Heskett wrote: >> There are no caps on a class D output, in fact the circuitry is isolated >> from as much capacitance in the load as possible with an inductor. >> Anything over 10 pfs or so just wastes switching power in the active parts >> and can lead to the destruction of the parts. >> >> For small power requirements, say 40 watts, Ramsey Electronics has a >> little 2" square, no heat sink required, 40 watt class D amplifier >> featured rather prominently on the front page of their latest catalog, not >> terribly expensive either. I would think, thinking as a C.E.T., that a >> few tens of microhenries of choke isolation on each output lead would >> allow any number of them to be paralleled for enough power to build your >> vfd regardless. 10 of them for a 400 watt vfd would probably still be >> cheaper than a commercially built vfd of that same power level. > >The switching waveform would have to be synchronized for this to >work, I would think. Otherwise, the inductors would need to be >much larger, or the amps would be fighting each other. > >Jon
I'm assuming sufficient isolation inductance so that it would be several hundred ohms impedance at the switching frequency which is probably in the 50 to 150 khz range, so it would be essentially a very small problem if the individual chips on each board weren't synched. As long as the inputs for a given phase are paralleled, it shouldn't be a problem. Not having one of them in my hot little hands though, I have NDI how much inductance is on each board but it would surely be nearly sufficient just to protect any piezo tweeters that may be attached. Looking it up, I see that its designed to run on the common 18-0-18 volt transformer, bridge rectified and filtered, and that the 40 watts in mono bridged mode is into a 4 ohm load. Most vfd's I'd assume run at higher voltages, so the additional weight of a suitable stepup transformer would need to be included in the budget. Ed Nisley (I think) wrote an article in Circuit City a few issues back on how to calculate and wind almost any sort of a transformer. The only problem I'd have with that would be iron saturation when the operating frequency goes low. With the core saturated, the iron effectively goes away leaving very little inductance to control the currents rate of rise vs time. This is exactly the same effect that destroys the H-scan circuitry in a crt monitor if the H-scan rate is forced below the makers minimum rated speed. The output transistor has been known to blow the epoxy off its top as it fails, its that violent. And of course when it fails shorted, several other $10 parts usually follow suit in a vain attempt to save a 50 cent fuse. :) Now, if we had motors designed for vfd drive that ran directly on that 50 volt peak to peak power from one of these, then it would be quite a bit more usable. OTOH Jon, my experience with vfd's is confined to the headwheel drivers in 40 year old 2" quadruplex broadcast VCR's. The operating frequency was nominally (+ - 5%) 400 hz, and the 3 phase driven motors were in the wheel and ran at 14,400 rpms. Later versions of these machines had amps that were good for several kilowatts of power for the first second, and were capable of bringing a stopped headwheel to within 20 microseconds of the correct rotational position within 400 milliseconds of the start signal, for a time base corrected and locked video signal less than half a second after the operator hit the start button. The ones I maintained back then were slower, needing 3.5 to 4 seconds to 'get locked' & were usually started 5 seconds or more before 'first video' in order for them to have time to lock up before the video was switched to the on air circuit. I suspect that much of what we know today about vfd's was discovered by ampex in developing those machines all those years ago. At one point, there were probably over 3000 of them in use at tv stations just here in the states. So it is a rather narrow field. Ampex would have cheerfully killed for such a circuit back them, to replace the drawers full of TO-3 transistors they used as analog amplifiers, and which required several tons of HVAC per machine just to get rid of the estimated 25kw power consumption of one of the AVR-1's when it was running. :) We've come a long way Baby! -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) You will be married within a year, and divorced within two. ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
