 |
|
|
 |
 |
|
|
[reads from bottom up..] DonW, Thanks! I needed that! Haven't built a thing in a couple of years in the way of a power supply. When I worked with them at TESLAco, I was too "green" to produce much. Later, working other places, the (then active) patents and resultant licensing requirements got in the way, so I opted for Buck Fwds. and the like! Many have attempted the Boostbuck topologies, and maybe you will succeed with one before I do. In a way they are so simple and natural that engineers tend to discredit the claims about them, favoring the "latest and greatest" contrivances in the journals and E-zines. But if you stick to your guns, your design will be best in the long run! P.S. Careful about that "simulator," she'll have you meeting her dad before you can say diggeridoo.... Bill, Thats quite clever ! So RCd for ur energy transfer damping cap TC1 and then use something filterwise du rigeur for ur output TC2. You assume that when Vg input goes dead the R is still small enough to refill ur energy transfer cap each sw cycle (??)... and that D opens as the voltage bleeds off. Neat. Have u actually DONE this Bill ??? ((Lots of big objects flapping in the breeze though!!)). Ive read a few papers on CUK converters and with the above in mind they seem to be a better topology for the PFC system BUT Im hesitant because of the Volt stress on the switch and the lack of ZVT. I've read a paper by same Athalye/Maksimovic/Erikson "high performance Front End Converter for Avionics Apps" which has a good active voltage clamp to provide ZVT PFC in CCM. Seen it ? Would that improve the efficiency likewise on ur boostBuck ? Im going to cogitate on the matter. In view of ur help Bill Ill contribute the results to you for inclusion on ur site if it pleases you.. ..I like community design. Maybe we can all work on it .... ur "normal" beef seems to have aroused a few sceptics. Nothing douses the ardour of a self righeous sceptic like a dose of reality. Me Im just curious to settle some of my own "ghosts" - and my experience (minimal so far mind) has thus far demonstrated that these power thingies "evolve " - no matter how rigorous and over toilet trained the designer! Ill get on my simulator straight away! heh heh rgds DonW Don: The general idea is that if one needs large electrolytics for holdup, one may as well use them as damping capacitors for the two basic LC filters in the converter. If you are not familiar with shunt damping, see: http://www.boostbuck.com/MoreHangupsSafetyFirst.html The two LC filters referred to are explained at: http://www.boostbuck.com/ImpedanceConditionsSafetyFirst.html Basically, you need one LC to keep the switched waveforms out of the source, and one to keep it out of the load! Note that the damping resistor lowers the Q of the LCd filter to well below unity, safely limiting inrush.... Bill Hi Bill, My question is dumber than that Bill - I wonder where you can get ur storage so the BuckBoost can drain the storage to zero. It seems innefficient to me to leave all that headroom on the energy storage device - which is what I thought u were getting at. So where did u mean the storage device to be - not the coupling caps ???? As we speak Im building a BoIBB PFC prototype as per Chen/Maksimovic/Erikson fm Colorado Power. Ive got bugger all room for inductance and they seem to suggest a good way to get low L component loss and small sizes. Bit worried about the switch losses but Ill see about something with that after I demonstrate the effective reduction in heat lost thru inductors. I need to get a lower voltage o/p - 200Vdc to feed a type of dPhase FBrdge. You tried the interleaved BuckBoost topol Bill - didnt see much mention ? Always liked CUK & Sepics tho never got game enough to really get into them - there was never enough time. But thats life...I REALLY like ur distinction between the engineer & the inventor. Sounds like you've been both. Rgds DonW Don: Main problem with the SEPIC is output noise. Spikes and ripple tend to "bounce off" the leakage inductance of the output caps. The Boostbuck topologies all feature an inductance between the switch and the load! These block switch noise while the output caps shunt it away from the load. Bill Your Name: Don Wiliams Company Name: Clearblue Software Pty Ltd Address: 11 Weroona Ave City: Woodford, NSW State: CO Zip: 2778 Country: Australia Phone: 61 2 47 587 985 Fax: E-mail: donw@clearblu.net Comments: Hi there "boostbuck" man.. You make an interesting claim that with the CUK/boostbuck PFC tolopology you can make the troublesome dV margin required for holdup period the same as the total voltage of the energy storage cap....?? How so ? Are you saying you make the transfer CAPs humungous ? Or at least big enough to feed the holdup time ? Then dont we run into the old inrush problem we tried to get rid of first up? Im just getting to grips with some new tolopoligies here ...sorry if its a dumb question. Rgds DonW I was always interested in ripple steering and SEPICs but always worried about the Volts stress on the switches. Have looked at all the "bridgeless" PFC stuff and with cheap transformers like the "flattransformer" lean toward distributed magnetics rather than integrated. Cascaded stuff like Unitrode publish is OK but for multi-output sharing seems uneconomical. So Im reintegrating the topology issue. Bit like Zen gardening really. I note some of our friends have gone into deep poetry ! | Further Beef | Background | Tips | Engineering vs. Science | | Isolation of the Cuk Converter | Historical Perspective | | Return Home | Safety First | Lesson in Ethics | The Consequences of Bad Design | Real World Applications | Deals | e-mail me! | |
||
 |
 |
