Searching \ for '[EE] power supply transient droop' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power/actodc.htm?key=power
Search entire site for: 'power supply transient droop'.

Exact match. Not showing close matches.
PICList Thread
'[EE] power supply transient droop'
2016\10\13@183305 by Gordon Williams

flavicon
face
I have a boost converter that I am switching on with a microcontroller ..  The draw on the power supply goes from about 50mA to 2A and the power supply drops about 1.3 volts for 10ms during turn-on.  Power supply is 5V.

During the power dip the voltage goes below the voltage allowable for the microcontroller per the data sheet.  The LEDs also flicker which I don't like.

I tried adding more capacitance at the microcontoller 330uF and this only improved it slightly by about 100mV.

I thought of adding a resistor or diode in the power line to the the microcontroller to see if that would improve things with some bulk capacitance at the micro.  Is there a standard way to handle this to get over the short low voltage period?

Gordon Williams


-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2016\10\13@190840 by IVP

face picon face
part 1 644 bytes content-type:text/plain; charset="iso-8859-1" (decoded quoted-printable)

> Is there a standard way to handle this to get over the short
> low voltage period?

Hi Gordon,

I have a couple of PIC-based circuits for driving clock solenoids
which also have moments of high current demand. To save costs
I use just a small transformer to charge capacitors from which the
current dump comes. The recharging of those capacitors depresses
the PSU for a short time, well below the PIC's minimum Vcc

I tried a resistor + reservoir cap to isolate the PIC from the current
demand but it wasn't enough. The attached circuit using steering
diodes works for my application

Joe

part 2 910 bytes content-type:image/gif; name="diode_isolation.gif" (decode)


part 3 181 bytes content-type:text/plain; name="Certification"
(decoded base64)

-----
No virus found in this message.
Checked by AVG - http://www.avg.com
Version: 2016.0.7797 / Virus Database: 4664/13204 - Release Date: 10/13/16
part 4 197 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2016\10\13@194325 by stephen.forrestn/a

flavicon
face
Hi Gordon -

It sounds like your boost supply is powered from the regulated micro supply.. Can you power it from the unregulated supply? What is the unregulated supply? Is that what is actually drooping? What sort of regulator is the 5V reg anyway?
It's amazing how one question cascades isn't it :o)

Stephen

{Original Message removed}

2016\10\13@195710 by Gordon Williams

flavicon
face
Thanks for the schematic.  That is what I was thinking and will give it a try.

Gordon



On 16-10-13 07:07 PM, IVP wrote:
{Quote hidden}

-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2016\10\13@200112 by Gordon Williams

flavicon
face
Both micro and boost converter come from the same regulated supply.  It is just during the turn-on transient that the supply can't act fast enough and the voltage drops momentarily.

Gordon



On 16-10-13 07:42 PM, spam_OUTstephen.forrestTakeThisOuTspamagilent.com wrote:
> Hi Gordon -
>
> It sounds like your boost supply is powered from the regulated micro supply. Can you power it from the unregulated supply? What is the unregulated supply? Is that what is actually drooping? What sort of regulator is the 5V reg anyway?
>
> It's amazing how one question cascades isn't it :o)
>
> Stephen
>
> {Original Message removed}

2016\10\13@200128 by RussellMc

face picon face

As ever, a circuit diagram would be a REALLY good idea, as would any
available specs on components used.

1. If not present, try adding a small cap (1 nF range) across the upper
resistor of the feedback divider. This should improve transient
performance. (And may do "other things" too.

2. Larger cap at the INPUT to the converter?


3. More maybe  after above tried.

​​

Russell
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2016\10\13@205852 by Brent Brown

picon face

As well as diodes & caps, look at feasibility of reducing the inrush current. If it can be reduced without too many negative effects that should make the problem smaller and easier to solve (or give more margin for a given solution). E.g. add series R, or see if "soft start" can be implemented with your boost converter.






-------- Original message --------
From: Gordon Williams <.....gwilliamsKILLspamspam@spam@ncf.ca>
Date: 10/14/16  1:01 PM  (GMT+12:00)
To: "Microcontroller discussion list - Public." <piclistspamKILLspammit.edu>
Subject: Re: [EE] power supply transient droop

Both micro and boost converter come from the same regulated supply.  It
is just during the turn-on transient that the supply can't act fast
enough and the voltage drops momentarily.

Gordon



On 16-10-13 07:42 PM, .....stephen.forrestKILLspamspam.....agilent.com wrote:
> Hi Gordon -
>
> It sounds like your boost supply is powered from the regulated micro supply. Can you power it from the unregulated supply? What is the unregulated supply? Is that what is actually drooping? What sort of regulator is the 5V reg anyway?
>
> It's amazing how one question cascades isn't it :o)
>
> Stephen
>
> {Original Message removed}

2016\10\13@211953 by James Cameron

flavicon
face
Had this once.  The transient was pulling energy out of the capacitor
before the regulator that powered the micro, so the regulated output
also collapsed.

Adding more capacitance on the micro side did not work, because the
regulator was happy to feed that energy backwards to the transient
load.  I think; was only using a swinging needle meter back then.

In the end I hid my regulator and input capacitor behind a diode, and
made sure I had enough capacitance to survive the transient.

Also used used PWM to soft-start the load, which was a 12V 2A nominal
incandescent lamp.

quozl.linux.org.au/tad/98-552/index.html
(ugh, shame, very early work)

-- James Cameron
http://quozl.netrek.org/
-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2016\10\13@212345 by RussellMc

face picon face

I realise  I read the question wrong (skimmed too fast & thought I'd
understood ?)

I was of course thinking about a dip in boost converter output and you were
talking about Vin dipping.

As before, a circuit AND context would help.

WHY is the boost converter loading the supply so much?
What is the supply capacity?
What is the boost converter load?
What is boost Vout?

Death by a thousand cuts Q&A is a pleasant enough way to while away one's
time, but far from efficient, usually.

The boost converter may be starting into a short circuit. Or not. You
probably know. We don't.

Vout_boost may be not so much higher than Vin that precharging the load to
5V may make a vast difference. Or not. Or may already be done. Or may not
be acceptable. You probably know. We don't.

The boost converter IC / circuit / specific arrangement may be amenable to
f\soft start easily and cheaply. Ot\r not. Or it may alraey be done. Or ...
.

I could go on a while longer on things we don't know that you probably do
and which would be useful for us to know. Or not.

More info would be nice.



        Russell


On 14 October 2016 at 12:00, RussellMc <EraseMEapptechnzspam_OUTspamTakeThisOuTgmail.com> wrote:

{Quote hidden}

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2016\10\13@212912 by RussellMc

face picon face

On 14 October 2016 at 10:33, Gordon Williams <gwilliamsspamspam_OUTncf.ca> wrote:

> I have a boost converter that I am switching on with a microcontroller
> .  The draw on the power supply goes from about 50mA to 2A and the power
> supply drops about 1.3 volts for 10ms during turn-on.  Power supply is 5V.
>
>
2A is the ​startup peak? / 2A is the new steady state with boost on?
PSU rating is?
If 2A is new steady state, what is the peak at startup? (eg 0.01 Ohm in psu
lead with scope across it will tell you).

TOO MUCH boost OUTPUT capacitance MAY be the problem.
ie - charging the cap may make load too high.


​           R
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2016\10\13@222546 by Harold Hallikainen

flavicon
face
You can often soft start a boost converter by putting a capacitor across
the top feedback resistor.

Harold



-- FCC Rules Updated Daily at http://www.hallikainen.com
Not sent from an iPhone.
-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2016\10\13@223345 by Gordon Williams

flavicon
face

So, the boost converter is a XL6009 based module
http://www.ebay.ca/itm/XL6009-DC-DC-Adjustable-Step-up-boost-Power-Converter-Module-Replace-LM2577-/130987148386?hash=item1e7f714462:g:uNEAAOSwXeJXdiqQ

I'm using the enable pin to turn it on and off (off means that the 5V
input - diode drop shows on output).  5V @ 2A in and 22V out.

http://www.datasheetspdf.com/datasheet/XL6009.html

The caps on mine are different than in the picture.  Mine has 220uF
input, 100uF output.  Datasheet shows 47uF input 220uF output typical
application. Picture shows 220/220uF

Lab PS being used 5A output.  I've also used another PS and had much the
same droop

I haven't measured the input rush.  The XL6009 has soft start and
internal switch over current protection is 4A.

I think that I need to try the diode and cap to isolate the uC during
the droop period.

Gordon Williams



On 16-10-13 09:28 PM, RussellMc wrote:
{Quote hidden}

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2016\10\14@114119 by Gordon Williams

flavicon
face
I tried the Schottky and cap and it is not going to work unless I put a 1000uF cap in there.  The droop (10ms) too long and current (50mA) is too high.  And I use up .3V headroom due to the Schottky.

Time for plan B.  Increase the voltage into the device and run the uC off of a separate voltage regulator.

Thanks,

Gordon Williams



On 16-10-13 10:33 PM, Gordon Williams wrote:
> I think that I need to try the diode and cap to isolate the uC during
> the droop period.

-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2016\10\14@205530 by Art

flavicon
face
OMG, that's brilliant Harold!

A

On 10/13/2016 10:25 PM, Harold Hallikainen wrote:
> You can often soft start a boost converter by putting a capacitor across
> the top feedback resistor.
>
> Harold
>
>
>

-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2016\10\15@062359 by RussellMc

face picon face

On 15 October 2016 at 03:41, Gordon Williams <KILLspamgwilliamsKILLspamspamncf.ca> wrote:

> I tried the Schottky and cap and it is not going to work unless I put a
> 1000uF cap in there.  The droop (10ms) too long and current (50mA) is
> too high.  And I use up .3V headroom due to the Schottky.
>
> Time for plan B.  Increase the voltage into the device and run the uC
> off of a separate voltage regulator.
>

​That sounds like plan G to me.
Or plan C at best.

As above
> TOO MUCH boost OUTPUT capacitance MAY be the problem.
> ie - charging the cap may make load too high

We have not been given all the available information, but from what we have
it seems inlikely that the problem needs tonbe solved by crute force
avaoidance of the energy deficiency glitch, tather than fixing the cause.

If you calculate the energy required to charge the cap you find it's
substantially lower than the energy in the glitch. The converter is
starting badly and drawing current whu\ich does nit seem to get to the
output. Feeding the cap energy at a rate that does not crowbar a lab supply
sounds likely to be feasible.

      Russell
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2016\10\15@065923 by IVP

face picon face
> Time for plan B

Maybe a PTC thermistor ?

I think every CRT TV / monitor had one, between the mains on/off
switch and PSU input. Usually an easily identifiable black disk

Joe


-----
No virus found in this message.
Checked by AVG - http://www.avg.com
Version: 2016.0.7858 / Virus Database: 4664/13213 - Release Date: 10/15/16

-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

More... (looser matching)
- Last day of these posts
- In 2016 , 2017 only
- Today
- New search...