X-Mailer: Mozilla 4.79 [en] (Win98; U)
Subject: Re: 87% All that vector calculus paid off
References: <email@example.com> <3E06A99F.776F9A81@ieee.org> <3E07486E.424879B1@ieee.org> <951O9.102279$4W1.firstname.lastname@example.org>
NNTP-Posting-Date: Wed, 25 Dec 2002 22:50:26 GMT
Organization: AT&T Broadband
Date: Wed, 25 Dec 2002 22:50:26 GMT
Harry Dellamano, analog, Chris Carlen wrote:
>> Nope. I was thinking of peak current control with a (nearly) constant
>> off time to let the drain voltage ring down to and *be*at*zero* when the
>> switch is turned on again. This method of control yields nearly lossless
>> switching (there are no f*C*V^2 losses) and avoids the source of stair
>> stepping altogether.
>> When the load is light, the peak current commanded by the error amplifier
>> will be small, and the ramp up time to achieve that peak will be quick.
>> Thus, frequency will be highest at light load (but it will never exceed
>> the natural ringing frequency, obviously). There are some ICs out there
>> that are designed for this mode of operation, or standard ones, like the
>> TC3842, can be made to work this way with a little helper circuitry.
>>> Thanks for the input.
>> Thanks for the interesting problem.
>>> Good day.
>> Yes, each day with something new to learn or an some interesting puzzle
>> to solve usually is. -- analog
> Analog, I mentioned to Chris in an earlier thread that it should be done
> with ZVS (Zero Voltage Switching) for better efficiency, no right side
> pole, lower EMI and just pretty pictures. The TC384N series is a way to
> go. Basically it is peak current control with constant off time.
As far as I know the TC384x series of ICs are single output, fixed
frequency, peak current mode control, not constant off time - although
with extra circuitry they can be made to work that way. If you were to
supply me the exact part number I would be happy to search for it on TI's
> I bet I can design it with less components than you, maybe. If you show
> me yours I will show you mine.
My, but you Brits are a cheeky bunch. Ok, then. For starters I would
ascertain some basic design information such as the input voltage range
required, the output voltage and current ranges required including
abnormal loads, and efficiency, noise, ripple and EMI requirements. It
seems keeping the magnetics as simple as possible is very important to
Chris, but if it were my project, I would probably go to a true flyback
or a push-pull dc-to-dc converter, depending on what Chris answered about
the project requirements.
But if we were to stick with Chris's basic boost design then I would
probably use an LM339 quad comparator and a few discrete components to
do the job (variable frequency, peak current control with zero voltage
switching - oh, and don't forget to add active turn-off dv/dt limiting
to decimate EMI generation).
> But the first thing I would do is trash that plug board. No good for
> high speed power. Each high current loop must be reduced and controlled
> maybe with a small ground plane for the control section. I tend to use
> Vector board with pins to hold the critical nodes. The control section
> can be wirewrapped if you are careful. Know your high dV/dT and dI/dT
> nodes. Also 50KHz is awfully slow, 100KHz may be better.
I, too, would use perf board. -- analog