dc/dc converter 12/200V

Jim Thompson · Apr 19, 2005

On Tue, 19 Apr 2005 16:19:59 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Tue, 19 Apr 2005 13:40:37 -0700, Jim Thompson
thegreatone@example.com> wrote:

On Tue, 19 Apr 2005 12:16:57 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

[snip]

Gosh, Jim, you're getting crabby. Of course I understand it; I did
point out, somewhere way up above, that the base drive width had to be
adaptive, perhaps with current sensing, to keep the energy per zap
constant.

Your insistance that I don't understand how it works is surprising.

John

Me? Crabby? I thought maybe you had one too many Hefe Weizen ;-)

No. My spicy stir-fry chicken demands one of those big Kirins.

For lunch, we had Steak Kew and shared a bottle of Chalone Chardonnay
;-)

Way back I DID explain the control system had "smarts".

And you kept insisting it wasn't "production".

Well, I did ask. Are race cars "production"?

John

480)460-2350 | |
| E-mail Address at Website Fax

Frithiof Andreas Jensen · Apr 20, 2005

"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in message
news:rjea61pr8a427q52g1fpc3njhp994dv6a2@4ax.com...

This circuit is cute and clever, and one could certainly build one and
get it to work (provided the base drive timing assured full energy
transfer at min cranking voltage and didn't fry things at max), but I
don't think it's the way people would do this in production.

It's exactly the kind of thing that makes it into production - because Low
Unit Cost beats *Anything*.

In this case the controller would consist of between Three to six
Transistors and *maybe* a reference (probably globbed up in some Epoxied
custom assembly, the better to gouge the spare parts buyers). You would
spend more on components just to get a uC to *survive* - and they you
haven't even done the code yet!

My first insight was when I got a "Philips application notes" Book from when
that was on paper: Hardly an IC in any of it, many, many discretes though -
I especially liked the entirely discrete switcher for charging a shaver; to
keep EMI down, the design stopped the switcher whenever the input rectifier
diodes conducted saving a filter.

So, yes, Clever does Make it! Fat and Bloated designs almost* only thrive
in Millitary applications - where cost is no object, there are people to
nurse the equipment 24/7, it takes years to get permission to solve known
problems, and only a few thousands are ever made.

*)

Weelll, Telecoms core also - big on management, mostly *useless* in it's
byzantine complexity which will induce any sane person to leave it well
alone, where maybe 60MB of valuable memory and Gigs of Hard Disks are
squandered to provide "management" to maybe 6 MB of "Job" - Oh and we need
an OS to run Java too because the management software is written in this -
and the complexity of the management drives the requirements for ever more
complex "mangement solutions" to software realiability.

"The Networks" tried to escape by introducing "All IP Networks" but "they"
did not achieve escape velocity and are now being pulled into the ground
overburdened with garbage software attempting to transmute an IP network
into an ATM network.

Jim Thompson · Apr 20, 2005

On Wed, 20 Apr 2005 11:17:08 +0200, "Frithiof Andreas Jensen"
<frithiof.jensen@die_spammer_die.ericsson.com> wrote:

"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in message
news:rjea61pr8a427q52g1fpc3njhp994dv6a2@4ax.com...

This circuit is cute and clever, and one could certainly build one and
get it to work (provided the base drive timing assured full energy
transfer at min cranking voltage and didn't fry things at max), but I
don't think it's the way people would do this in production.

It's exactly the kind of thing that makes it into production - because Low
Unit Cost beats *Anything*.

Yep.

In this case the controller would consist of between Three to six
Transistors and *maybe* a reference (probably globbed up in some Epoxied
custom assembly, the better to gouge the spare parts buyers).

Nope. It wasn't too many transistors but it WAS an ASIC in a ceramic
DIL. Pickup processing (analog), BandGap and temperature tracking and
that sort of stuff. Since emitter current was measured, there was
also a circuit to calculate (also analog) collector current, as the
point of shut-off.

You would
spend more on components just to get a uC to *survive* - and then you
haven't even done the code yet!

[snip]

480)460-2350 | |
| E-mail Address at Website Fax

John Larkin · Apr 20, 2005

On Wed, 20 Apr 2005 11:17:08 +0200, "Frithiof Andreas Jensen"
<frithiof.jensen@die_spammer_die.ericsson.com> wrote:

"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in message
news:rjea61pr8a427q52g1fpc3njhp994dv6a2@4ax.com...

This circuit is cute and clever, and one could certainly build one and
get it to work (provided the base drive timing assured full energy
transfer at min cranking voltage and didn't fry things at max), but I
don't think it's the way people would do this in production.

It's exactly the kind of thing that makes it into production - because Low
Unit Cost beats *Anything*.

Well, is it in production? Is it low cost?

In this case the controller would consist of between Three to six
Transistors and *maybe* a reference (probably globbed up in some Epoxied
custom assembly, the better to gouge the spare parts buyers). You would
spend more on components just to get a uC to *survive* - and they you
haven't even done the code yet!

In a modern car, the control would use up about 2% of the capacity of
the engine control computer, which is computing the advance anyhow.

John

Frithiof Andreas Jensen · Apr 21, 2005

"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in message
news:8kqc6111egnftkq4nsqq3c0cgscmpnn5rb@4ax.com...

In a modern car, the control would use up about 2% of the capacity of
the engine control computer, which is computing the advance anyhow.

Ah - But we were talking about a Motorcycle initially ...

PS:

I think them modern cars could do with some Management; there seems to be
ressources in need of wasting ;-)

Jim Thompson · Apr 21, 2005

On Wed, 20 Apr 2005 07:47:00 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

[snip]

In a modern car, the control would use up about 2% of the capacity of
the engine control computer, which is computing the advance anyhow.

John

"engine control computer", the most commonly failing component in GM
vehicles in AZ ;-)

BTW, Advance isn't "computed", it's dithered until various sensors
report stoichiometric and/or onset-of-ping.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice

480)460-2350 | |
| E-mail Address at Website Fax

John Larkin · Apr 21, 2005

On Thu, 21 Apr 2005 07:22:59 -0700, Jim Thompson
<thegreatone@example.com> wrote:

On Wed, 20 Apr 2005 07:47:00 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

[snip]

In a modern car, the control would use up about 2% of the capacity of
the engine control computer, which is computing the advance anyhow.

John

"engine control computer", the most commonly failing component in GM
vehicles in AZ ;-)

More than water pumps? Batteries? I've replaced lots of those, but no
ecc's so far.

BTW, Advance isn't "computed", it's dithered until various sensors
report stoichiometric and/or onset-of-ping.

That sounds sort of like computing to me.

John

Jim Thompson · Apr 21, 2005

On Thu, 21 Apr 2005 07:45:28 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 21 Apr 2005 07:22:59 -0700, Jim Thompson
thegreatone@example.com> wrote:
[snip]

"engine control computer", the most commonly failing component in GM
vehicles in AZ ;-)

More than water pumps? Batteries? I've replaced lots of those, but no
ecc's so far.

I'm talking infant failures... within weeks of taking possession of
the new vehicle.

BTW, Advance isn't "computed", it's dithered until various sensors
report stoichiometric and/or onset-of-ping.

That sounds sort of like computing to me.

John

480)460-2350 | |
| E-mail Address at Website Fax

John Larkin · Apr 21, 2005

On Thu, 21 Apr 2005 08:12:45 -0700, Jim Thompson
<thegreatone@example.com> wrote:

On Thu, 21 Apr 2005 07:45:28 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 21 Apr 2005 07:22:59 -0700, Jim Thompson
thegreatone@example.com> wrote:
[snip]

"engine control computer", the most commonly failing component in GM
vehicles in AZ ;-)

More than water pumps? Batteries? I've replaced lots of those, but no
ecc's so far.

I'm talking infant failures... within weeks of taking possession of
the new vehicle.

BTW, Advance isn't "computed", it's dithered until various sensors
report stoichiometric and/or onset-of-ping.

That sounds sort of like computing to me.

John

UNNATURAL digital implementation of a natural analog process ;-)

...Jim Thompson

If you just remind yourself that all those microprocessors are really,
inside, just a bunch of overdriven, CMOS push-pull class-A linear amp
stages arranged in feedback loops, you'll feel a lot better about
them.

John

Jim Thompson · Apr 21, 2005

On Thu, 21 Apr 2005 08:46:49 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 21 Apr 2005 08:12:45 -0700, Jim Thompson
thegreatone@example.com> wrote:

On Thu, 21 Apr 2005 07:45:28 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 21 Apr 2005 07:22:59 -0700, Jim Thompson
thegreatone@example.com> wrote:
[snip]

That sounds sort of like computing to me.

John

UNNATURAL digital implementation of a natural analog process ;-)

...Jim Thompson

If you just remind yourself that all those microprocessors are really,
inside, just a bunch of overdriven, CMOS push-pull class-A linear amp
stages arranged in feedback loops, you'll feel a lot better about
them.

John

Of course I do. As the Absolute Master God of Analog I am often
called upon to fix issues with "digital" circuits, primarily at
Motorola/ON-Semi and Intel, but recently also at Microchip ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice

480)460-2350 | |
| E-mail Address at Website Fax

Jim Thompson · Apr 23, 2005

On 23 Apr 2005 14:27:47 -0700, "kell" <kellrobinson@billburg.com>
wrote:

The commercially available CDI units made by companies like MSD do not
fire only once per combustion cycle.
The spark you get from dumping a cap into the primary of an ignition
coil lasts only a fraction as long as conventional inductive spark.
This can have adverse effects on driveability. If there is a lean spot
in the mixture, you might not get combustion; and modern engines run
lean.
MSD's product and others fire multiple times per ignition, when the
rpm's are low enough to allow it. At higher rpm's, they go to a single
spark.
I guess for practical reasons OEM ignition systems don't use CDI, at
least on cars; on smaller engines, yes.
Modern ignitions have gotten away from the old-fashioned degree dwell,
where the coil current is limited by primary resistance. You see coils
now with just a fraction of an ohm primary resistance and
computer-controlled dwell.

480)460-2350 | |
| E-mail Address at Website Fax

dc/dc converter 12/200V

Jim Thompson

Guest

Frithiof Andreas Jensen

Guest

Jim Thompson

Guest

John Larkin

Guest

Frithiof Andreas Jensen

Guest

Jim Thompson

Guest

John Larkin

Guest

Jim Thompson

Guest

John Larkin

Guest

Jim Thompson

Guest

Jim Thompson

Guest

Welcome to EDABoard.com

Sponsor

Online statistics

Forum statistics

dc/dc converter 12/200V

Jim Thompson

Guest

Frithiof Andreas Jensen

Guest

Jim Thompson

Guest

John Larkin

Guest

Frithiof Andreas Jensen

Guest

Jim Thompson

Guest

John Larkin

Guest

Jim Thompson

Guest

John Larkin

Guest

Jim Thompson

Guest

Jim Thompson

Guest

Log in

Welcome to EDABoard.com

Sponsor