From: Chris Carlen
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
Subject: Re: 87% ... Design requirements are here
References: <email@example.com> <3E06A99F.776F9A81@ieee.org> <3E07486E.424879B1@ieee.org> <951O9.102279$4W1.firstname.lastname@example.org> <3E0A3631.4A5C4AC5@ieee.org>
Date: Thu, 26 Dec 2002 21:40:29 GMT
NNTP-Posting-Date: Thu, 26 Dec 2002 13:40:29 PST
Organization: EarthLink Inc. -- http://www.EarthLink.net
Harry Dellamano wrote:
>>Here are the constraints I am working with. If you want to work with
>>these parameters, then I would be able to contribute to testing. You
>>draw, I build and test. Final design should be "open-source" hardware
>>(freely offered and able to be copied and distributed as long as
>>original contributors' names are included.) If you don't want to do it
>>this way, then I can't participate in testing, and will have to make my
>>own device on my own, using only conceptual guidance from this forum.
>>11-25VDC input voltage range. 12W maximum output power. 90-350V
>>adjustable output voltage. Circuit should still operate down to 8V
>>input, but at reduced output power. If you don't like this, then forget
>>this 8V input functioning requirement.
>>Prefer a boost or flyback topology, isolation of output voltage from
>>input is *not* needed, but if a flyback "transformer" rather than plain
>>inductor is better for meeting design requirements or simplifying
>>circuitry, then use it.
>>Regulation and output ripple specs are not critical, you decide. Must
>>tolerate short circuited outputs, and open circuited outputs. A default
>>minimum load resistor or current sink is Ok, as long as it works over
>>the entire 90-350V output and can be disconnected if desired.
>>85% minimum efficiency (default load power dissipation may be included
>>in efficiency calcs.) No EMI requirements, for hobby use. If any extra
>>circuitry is needed to reduce EMI, it should be easy to and obvious how
>>to strip it from the basic system if it is not desired, without breaking
>>Preferrably use an E19/8/5 standard core size, gapped to whatever is
>>needed, using 3C90 or 3C94 material. Max switching frequency 200kHz. I
>>have these cores in these materials on hand, and some small 3F3 and 3E5
>>toroids on hand if needed for gate drives, but these shouldn't be needed
>>for a flyback anyway.
>>I will be leaving for Thailand on Dec 29, and will return the week of
>>Jan 18. If you can wait until then, I'll start testing when I return.
>>Otherwise, do what you want, I'll be interested to see the results when
>>Christopher R. Carlen
> Sounds ok to me. So this is a non isolated dc/dc converter. You even give
> us a schedule when it must be completed to make it seem like a real task.
> Hopefully we can review all the designs by Jan 18 and have one consolidated
> design for you to start testing.
BTW, if it can put out more than 12W, that is Ok, but 12W min over the
input voltage range. I have already pushed 16W through these cores
(flyback topology at 150kHz) without distress, so this should be no problem.
I have a long list of parts waiting to order, so when I come back I can
add whatever is needed to the list and be ready to start testing a week
I have an analog 20MHz scope, a Tek TDS3014 100MHz 4-ch DPO scope (I can
put screenshots of test results on my website for review), but only 300V
probes. I plan to order a probe good to 600V for this kind of work. I
really want a differential probe, but can't afford that :-( Also a
function generator and a dual adjustable 1.2-30V 5A power supply.
I can handle about $50-$100 worth of expense for parts I don't already
have and that aren't already on my list for general stockpile building.
But I have the following (in addition to the E19/8/5-3C90/3C94 cores
that I mentioned):
6 of IRF511
15 of IRFI520N (these could be a nice choice, with isolated tabs, may
reduce drain capacitance to ground.)
3 of IRF730A
10 of IRF740
2 of IRF840
6 of 2SK2841
I like the heatsink equivalent to Wakefield 240, but if something
smaller is needed, I can deal with that.
30,28,26, and 22 ga. magnet wire.
HER105 diodes (will be ordering a bunch more of all types of diodes)
0.1, 0.22, 0.47, and 1.0uF 400V polyester or polypropylene film caps
Don't have many electrolytics, but I'm planning to stock up on some good
Panasonic FC and EB caps.
Current sense resistors: I am sorely lacking these, but plan to get a
bunch, as well as various load resistors.
Misc logic chips, general discretes, and other junk.
Oh, I prefer to order parts stocked by Digi-Key, Allied, Mouser, or
Newark. Other distributors that usually deal in high volumes are
something I try to avoid.
Once a design is settled and tested, I'll lay it out in 2-layers in
Eagle and put in on the web for review before committing to FR4. I
might make the layout combined with a small transformer rectifier and
3-terminal regulator PCB, but the DC-DC converter section would be
All resulting files and schematics I'll host on my web site with a list
of contributors (including my wife if she winds the coils! ;-)
> Have a great trip and if this NG is still talking to each other when you
> get back we should have a mature design.
Thanks. I think the NG will be fine. They always make up later.
Christopher R. Carlen
Suse 8.1 Linux 2.4.19