Maker Pro
Maker Pro

Multiplex your inductor

W

Winfield Hill

Jan 1, 1970
0
I missed this when it first came out in 2004 - TI
introduced a four-output SMPS chip meant for cell
phones, very small, and to even further save size,
it required only one small inductor. The secret?
Multiplex your precious inductor cycle-by-cycle,
among three outputs! Create +5, +20 and -18, all
from one 10uH inductor. It's TI's tps65120 to 24.

http://focus.ti.com/docs/prod/folders/print/tps65120.html

Stay happy.
 
J

Joerg

Jan 1, 1970
0
http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1051
MAX212, done in 1994. I'm sure someone beat me to it since the idea is
pretty simple. You are limited to discontinuous conduction converters.
The 212 is clockless (self timed) so that the inductor value isn't
critical. Also a very high PSRR internal reference to avoid a
reference bypass cap. You had to make these parts very simple for the
232 crowd. In the day when you had serial mice, you wish you had these
chip on your notebook mobo.


And as usual nearly unobtanium, a whopping 58 pieces left to be scraped
off the bottom of the barrel:

http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=MAX212

It's sad, they employ smart engineers like yourself and then seem to
waste so much of their creativeness by screwing up the logistics.

This is how it's done right:

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=296-17184-1-ND
 
L

legg

Jan 1, 1970
0
I missed this when it first came out in 2004 - TI
introduced a four-output SMPS chip meant for cell
phones, very small, and to even further save size,
it required only one small inductor. The secret?
Multiplex your precious inductor cycle-by-cycle,
among three outputs! Create +5, +20 and -18, all
from one 10uH inductor. It's TI's tps65120 to 24.

http://focus.ti.com/docs/prod/folders/print/tps65120.html

Stay happy.

The power-up and power-down sequencing looks scarey. You wouldn't want
too much interdependency between Main and Vgh, as is.

RL
 
J

Joerg

Jan 1, 1970
0
Granted the Maxim management can't manage, and the higher you go, the
worse it gets, but to be fair, who is doing 232 ports these days? And
if you are doing 232s, it is probably on a PCI card, so cheaper
solutions can be used.

232 is just about the only thing I am not so concerned about, but only
if there's 2nd source.

There was a follow on part from Maxim that muxed the inductor. Not
mine, so I can't pull the number out of my head, and the Maxim on-line
search is pretty poor if you want to find this kind of stuff.

Digikey sales are a bit strange. At time, the catalog reads like an
advert for the part vendor. Well, that is because the vendor pays
Digikey to run that stuff. As far as I know, the vendors eats the cost
because it is hard to mark up the parts to make the difference. I've
never been able to get someone to give me the straight dope about how
this is done.

I can't dis' TI. I wish they would buy Maxim and clean house. The
stock is still depressed due to the criminal activity (i.e. illegal
stock repricing), but eventually it should get back on par with LTC.


I was actually just thinking about buying LTC for our IRA. Healthy
fundamentals, they pay dividends, but the high P/E of 28 ain't my cup of
tea. Personally I'd never buy Maxim stock though, ever. Unless 100% of
top management is changed out _and_ I've seen at least two years of
consistent volume shipments.

TI would do better to get the stock when it's depressed. If the Maxim
board had any balls, they would have canned the illegal traders and
got on with the chip business instead of the ass protection business,
but they dragged things out, getting the stock delisted. The board is
a totally spineless group that is well paid to look the other way.

I've never looked, since I never design in MAX parts ;-)
 
J

Joerg

Jan 1, 1970
0
232 is just about the only thing I am not so concerned about, but only
if there's 2nd source.

I was actually just thinking about buying LTC for our IRA. Healthy
fundamentals, they pay dividends, but the high P/E of 28 ain't my cup of
tea. Personally I'd never buy Maxim stock though, ever. Unless 100% of
top management is changed out _and_ I've seen at least two years of
consistent volume shipments.

I've never looked, since I never design in MAX parts ;-)

I think analog is a good market. Not so many players. The winners
invent new playgrounds (phones, computers, portable devices, etc). LTC
to their credit has held onto bipolar technology longer than I
expected would be profitable. [You can't do LSI bipolar for a lot of
reasons.]

Until the criminal activity (my words, since they never admitted to
criminal activity), Maxim exceeded LTC in growth, though LTC was
always the more profitable company. ...


To me, growth means nothing if it's more or less artificial. If
potential customers repeatedly can't get product delivered then the
executives didn't properly execute.

... They have ROI up their with the
best software companies, and you know hardware (components) has a
greater energy barrier (barrier to market entry) than software.

Steve Jobs did a lot of that illegal stock repricing. You have a
billion dollars, so what is the big deal about being down $20 million
on paper? Idiot.

I've personally benefited from stock repricing, but did it legally.
You trade in the underwater shares for new shares, but reset the
clock. There isn't an engineer in the valley that doesn't know how
reprice an option legally.

Always depends on the company. The ones I worked for never did or
allowed such tricks. No re-pricing.
 
Top