Searching \ for '[EE]: Ready to through a reel of boost converters' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: piclist.com/techref/index.htm?key=ready+through+reel
Search entire site for: 'Ready to through a reel of boost converters'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Ready to through a reel of boost converters'
2003\06\27@104614 by A.J. Tufgar

flavicon
face
Hey everyone,
            I recently got a second revision of our PCB back from our
PCB house.  I through everything on the board and tried to fire it up
and as Murphy's law would have it did nothing.  :P

I traced it quickly to the power supply part of the circuit, which was
outputting .9V when it should have been 3.6V.

I took another PCB and soldered on only the power supply components.
No luck, although sometimes it gets really hot and outputs 6.6V??  :P

I thought ok an obvious blunder in making up the PCB, rechecked the
board many times looks ok and DRC checks out....

Took the meter to it all connections seem to go where they are suppose
to....

The only thing that has changed from this revision to the last are
widening the VCC and gnd, Using Tantalum instead of electrolytic and
using PCB mount battery clips instead of a separate battery holder.

I'm using Linear's LTC3400 and their standard standard circuit for the
power supply, it's worked in many projects before, but I'm at a total
loss on this one.  Anyone have any ideas I realize it's vague, but I
don't think it's just being temperamental, boost converters don't seem
to be that picky about layout.

Anyways, if anyone could offer a hand it would be more then GREATLY
appreciated.

Thanks in Advance,
Aaron

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2003\06\27@114633 by Tom Messenger

flavicon
face
At 10:45 AM 6/27/2003 -0400, you wrote:

Looking at the data sheet for the LT3400, the first thing that comes to
mind is the absolute specs for voltage on just about any pin are less than
6 volts.  So if you are producing more as you mentioned, there's some trouble.

Check your feedback network for correct values with an ohmmeter to catch
any mismarked resistors.

Boost circuits that overheat often indicate an inductor hitting saturation;
that's an area to check.

You mention changing trace sizes as well as going to a tantalum from an
electrolytic cap.  These things don't have much effect at low frequencies
usually... but I see from the data sheet that this part runs at 1MHz. How
does your pcb layout compare to the suggested layout in the data sheet? Got
short fat traces everywhere?

Check for a weak battery also.  The internal resistance goes up as the
battery gets discharged and could contribute to oscillation.

My limited experience in this area says that if there are no obvious
blunders, then the likeliest suspects are inductor saturation or feedback
circuit oscillation.

Let us know what you find.

Good luck!
Tom M.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2003\06\27@121755 by A.J. Tufgar

flavicon
face
Thanks Tom,
          The traces going to the inductor are as short and fat as
possible, about 20 mils away and 25 mil wide.  :)  Ground unfortunately
has to be carried the length of the board, as I discussed I'm now using
batter clips, but the trace is 25 mil wide.  The trace joining the two
resistors could be wider but this really carries almost zero current.
Resistors are 1M and 500K approx.  Traces to caps are short, but they
are on the underside of the board, but I used a large via to connect
them.

Inductor is recommended from the datasheet so I don't think it would be
saturating, I'm only drawing a couple mA also.

Thanks for your help,
Aaron

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2003\06\27@131540 by David Minkler

flavicon
face
Hi Aaron,

Any chance the Tants are installed backwards?  We had very similar
results on a co-workers circuit and that turned out to be the problem.

Dave

A.J. Tufgar wrote:
> I traced it quickly to the power supply part of the circuit, which was
> outputting .9V when it should have been 3.6V.
>
> I took another PCB and soldered on only the power supply components.
>  No luck, although sometimes it gets really hot and outputs 6.6V??  :P
> ...
> The only thing that has changed from this revision to the last are
> widening the VCC and gnd, Using Tantalum instead of electrolytic and

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2003\06\27@132955 by David VanHorn

flavicon
face
>
>Anyways, if anyone could offer a hand it would be more then GREATLY appreciated.

If you like, pop a pair in the mail to me, and I'll have a look.
It is probably something a little sneaky.
If it's not too involved, I won't even bill you. :)

I know what it's like to be in that position.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2003\06\27@164127 by Brian Aase

picon face
A.J. Tufgar wrote:
> I traced it quickly to the power supply part of the circuit, which was
> outputting .9V when it should have been 3.6V.
>
> I took another PCB and soldered on only the power supply components.
>  No luck, although sometimes it gets really hot and outputs 6.6V??  :P
> ...
> The only thing that has changed from this revision to the last are
> widening the VCC and gnd, Using Tantalum instead of electrolytic and

Do you have a scope available to you?  Is it showing about the same
waveforms as your previous working version?  Try looking for
subharmonic operation... layout changes can cause this for no obvious
reason sometimes.  Also, are you *sure* your inductor really is what
you think it is?  Do you have a means to measure its value, or to
observe the waveform of the current passing through it?

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

2003\06\27@165935 by gtyler

flavicon
face
Hi A.J.,
       Which voltage regulator are you using? Some are very sensitive to
output capacitor ESR and need a capacitor with an ESR in a defined range to
avoid instability. These are typically the low dropout types with dropouts
below about 1V as this means that they are using a pnp series pass element
and need the pole and zero created by the cap for stability. often a tant.
or ceramic cap will not work. Some manufacturers now offer regulators that
will work with any load cap. (look for "anycap" in google). I found it
better to use regulators like the LM1117 that do not give stability
problems. (we retrofitted 1000's of products for this reason)

George Tyler
{Original Message removed}

2003\06\30@110701 by A.J. Tufgar

flavicon
face
Well thank you for all your suggestions I still have have very little
luck.

I called linear and someone is suppose to be assisting me in find a
solution.  I've sent them my orcad files so hopefully they will be able
to find something.

I'll make sure to post to the list what I find.

Thanks all,
Aaron

--
http://www.piclist.com hint: To leave the PICList
spam_OUTpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu>

2003\06\30@133705 by A.J. Tufgar

flavicon
face
Just wanted to let the list to know I talked to linear and they told me
the problem is most likely cause by mechanical bouncing when the
battery is plugged in.  I can't really remember the whole theroy behind
it, but besically, the bouncing would cause the sw pin to see more then
6 volts(becuase of di/dt) and would cause a condition where the device
would never really settle to the proper voltage.

I guess I buy the story, but I have one of those "hmmmm, I dunno about
that feelings".  I'll have to try it and see.

Was a little preterbed also because I think the datasheet should make
some mention of this, it says in the datasheet "Typically, SHDN should
be connected to VIN through a 1M pull-up resistor".  But I guess it's
my fault for not thinking of debouncing.

I'll keep you posted,
Aaron

--
http://www.piclist.com hint: To leave the PICList
.....piclist-unsubscribe-requestKILLspamspam@spam@mitvma.mit.edu>

More... (looser matching)
- Last day of these posts
- In 2003 , 2004 only
- Today
- New search...