How to debug and fix voltage droop in SMPS

[Douglas Beeson]

Hi all,

I am trying to get a TI TLV61220 step-up voltage converter to work properly under load. This is a difficult part for me to work with as it is very small -- SOT23-6 -- and, I suspect, finnicky about layout. My conditions are: Input: 2 AA batteries = 1.8-3 V; Output: 5 V; load current: 20mA

I have set up a test platform on 1-sided, 1oz etched copper which is not exactly like the recommended layout, but I think I have avoided the most obvious problem of output current into the feedback path. For test purposes, I have soldered the part to a SOT23-6 -> DIP6 adapter board (also provided by TI) and soldered the pins of that adapter to my test board. Too much induction?

Unloaded, the circuit gives me the desired 5 V (5.1 actually..). But as soon as I load it with a single LED (6 mA), the voltage droops to 3.17 V. According to the spec sheet https://www.dropbox.com/s/kq0092xnxlozdeu/tlv61220.pdf?dl=0 , I should be able to provide 30-40 mA of current at 5V. I am using the recommended cap and inductor values.

Here is the circuit: https://www.dropbox.com/s/uu1zerdq7msfkpt/Beeson_TLV61220.pdf?dl=0

Where should I start looking for the cause and a fix?

Thank you!

doug

 

I have no idea of your skill level, so if I say someting too obvious don't take it personal.

First, how do you know the LED is only pulling 6 mA ? If it is a typical about 2 volt LED that would mean like a 500 ohm resistor in series with it. For testing actually you could just use resistors.

Another thing, to tighten up regulation you can change the top resistor to a lower value and use a couple of junction drips like across 1N400X diodes. As long as the temperature stays halfway within reason it shouldn't drift too much. If you need high precision on the output voltage though forget it.. Thermal comp will be a PITA.

I would be havin' a scope on that choke. See what happens to the duty cycle under different load conditions. Also with varying input voltages.

Also, another dummy one, if you are using batteries, are they maintaining the input voltage ? Batteries vary vastly in current capability and amp-hours. Ideally you should be using a bench power supply with a current meter on it.

I tellya, you can't trust batteries. got a batch in and while they work well in remote controls, in the digital camera they are only good for two or three shots. So if you are not assured of the proper input voltage, monitor that when applying the load, if you haven't. (that's what I mean by don't take offense, some people really would not have thought to do that)

Anyway, with the information given, that's all I got.

 

[Tim Wescott]

On Sat, 26 Sep 2015 20:05:08 -0400, Douglas Beeson wrote:

Hi all,

I am trying to get a TI TLV61220 step-up voltage converter to work
properly under load. This is a difficult part for me to work with as it
is very small -- SOT23-6 -- and, I suspect, finnicky about layout. My
conditions are: Input: 2 AA batteries = 1.8-3 V; Output: 5 V; load
current: 20mA

I have set up a test platform on 1-sided, 1oz etched copper which is not
exactly like the recommended layout, but I think I have avoided the most
obvious problem of output current into the feedback path. For test
purposes, I have soldered the part to a SOT23-6 -> DIP6 adapter board
(also provided by TI) and soldered the pins of that adapter to my test
board. Too much induction?

Unloaded, the circuit gives me the desired 5 V (5.1 actually..). But as
soon as I load it with a single LED (6 mA), the voltage droops to 3.17
V. According to the spec sheet
https://www.dropbox.com/s/kq0092xnxlozdeu/tlv61220.pdf?dl=0 , I should
be able to provide 30-40 mA of current at 5V. I am using the recommended
cap and inductor values.

Here is the circuit:
https://www.dropbox.com/s/uu1zerdq7msfkpt/Beeson_TLV61220.pdf?dl=0

Where should I start looking for the cause and a fix?

Thank you!

doug

If you're really using an electrolytic cap -- stop. Surface-mount
ceramic 10uF caps are vastly cheap these days, and are almost certainly
what's intended for this circuit.

A 'lytic will give you a horrible amount of ripple at the switching
frequency, and may well be confusing the regulator into thinking that
it's holding the set point just fine.

I haven't used this exact part but I've used ones like it and not had
trouble, so I'm not sure why it's giving you fits.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

 

[Tim Wescott]

On Sun, 27 Sep 2015 09:43:55 -0700, jurb6006 wrote:

I have no idea of your skill level, so if I say someting too obvious
don't take it personal.

First, how do you know the LED is only pulling 6 mA ? If it is a typical
about 2 volt LED that would mean like a 500 ohm resistor in series with
it. For testing actually you could just use resistors.

Another thing, to tighten up regulation you can change the top resistor
to a lower value and use a couple of junction drips like across 1N400X
diodes. As long as the temperature stays halfway within reason it
shouldn't drift too much. If you need high precision on the output
voltage though forget it. Thermal comp will be a PITA.

I would be havin' a scope on that choke. See what happens to the duty
cycle under different load conditions. Also with varying input voltages.

Also, another dummy one, if you are using batteries, are they
maintaining the input voltage ? Batteries vary vastly in current
capability and amp-hours. Ideally you should be using a bench power
supply with a current meter on it.

I tellya, you can't trust batteries. got a batch in and while they work
well in remote controls, in the digital camera they are only good for
two or three shots. So if you are not assured of the proper input
voltage, monitor that when applying the load, if you haven't. (that's
what I mean by don't take offense, some people really would not have
thought to do that)

Anyway, with the information given, that's all I got.

+1 on using a scope, if you have on.

If you don't have one, then even a cheap one will help. For this job you
probably want to look for an advertised bandwidth at least four times the
switching rate of the chip -- more is always better, or would be if it
didn't translate to more money.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

 

"A 'lytic will give you a horrible amount of ripple at the switching
frequency, and may well be confusing the regulator into thinking that
it's holding the set point just fine. "

I've seen shit like that. Oscillators will not oscillate, but other things will...

 

"If you don't have one, then even a cheap one will help. For this >job you
probably want to look for an advertised bandwidth at least four times >the
switching rate of the chip -- more is always better, or would be if >it
didn't translate to more money. "

Buddy of mine collects, buys and sells scopes. For like fifty bucks or so you should be able to get a Tenma/Elenco/Elcheapo scope that works just fine to 20 MHz, which is high enough for this no question.

We get them sometimes for cheap either not working or whatever, and then after that I have been calibrating them to a Tektronix 422, which is about the most stable and accurate scope in the world, within its limitations of course.

Bottom line is you do not need a new scope, and when learning, you want a cathode ray oscilloscope. (CRO) Even if you got a shit ton of electronics experience, if you don't with a scope, start with a CRO.

If you DO have a scope of course put it on that inductor to see what is happening. Take pictures and stick them up in Dropbox so we can see. (PLEASE use the public directory and not the fucking picture directory)

 

[Tim Wescott]

On Sun, 27 Sep 2015 18:47:33 -0700, jurb6006 wrote:

"If you don't have one, then even a cheap one will help. For this >job
you probably want to look for an advertised bandwidth at least four
times >the switching rate of the chip -- more is always better, or would
be if >it didn't translate to more money. "

Buddy of mine collects, buys and sells scopes. For like fifty bucks or
so you should be able to get a Tenma/Elenco/Elcheapo scope that works
just fine to 20 MHz, which is high enough for this no question.

We get them sometimes for cheap either not working or whatever, and then
after that I have been calibrating them to a Tektronix 422, which is
about the most stable and accurate scope in the world, within its
limitations of course.

Bottom line is you do not need a new scope, and when learning, you want
a cathode ray oscilloscope. (CRO) Even if you got a shit ton of
electronics experience, if you don't with a scope, start with a CRO.

If you DO have a scope of course put it on that inductor to see what is
happening. Take pictures and stick them up in Dropbox so we can see.
(PLEASE use the public directory and not the fucking picture directory)

I'm not sure that I agree that you want to start with a CRO, but then, I
haven't taught anyone.

If you're dealing with a lot of non-periodic signals, having a DSO is
essential: perhaps "best" would be a DSO that does a really good job at
emulating a CRO.

(I have an Agilent 546222 mixed-signal scope which has been nearly
perfect for me -- there are a few times on every project when I might
wish for more analog channels or more digital channels, but it's rare,
and I can usually work around them. For debugging deeply embedded
hardware and software, though, it's great.)

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

 

[Douglas Beeson]

On Sun, 27 Sep 2015 18:47:33 -0700 (PDT)
jurb6006@gmail.com wrote:

"If you don't have one, then even a cheap one will help. For this >job you
probably want to look for an advertised bandwidth at least four times >the
switching rate of the chip -- more is always better, or would be if >it
didn't translate to more money. "

Buddy of mine collects, buys and sells scopes. For like fifty bucks or so you should be able to get a Tenma/Elenco/Elcheapo scope that works just fine to 20 MHz, which is high enough for this no question.

We get them sometimes for cheap either not working or whatever, and then after that I have been calibrating them to a Tektronix 422, which is about the most stable and accurate scope in the world, within its limitations of course.

Bottom line is you do not need a new scope, and when learning, you want a cathode ray oscilloscope. (CRO) Even if you got a shit ton of electronics experience, if you don't with a scope, start with a CRO.

If you DO have a scope of course put it on that inductor to see what is happening. Take pictures and stick them up in Dropbox so we can see. (PLEASE use the public directory and not the fucking picture directory)

Thanks for the good advice. I actually did start with a 20MHz Tenma (got it free from a buddy) a few years ago but have since moved up to a Tektronix entry-level digital scope. I am really enjhoying it.

Here is a scope capture across the inductor (between BAT and SW pins). I see spikes and some ringing. I'm not sure whether this is "normal".

https://www.dropbox.com/s/g1x02uxmrtsoseb/4.7uH%20inductor.JPG?dl=0

I also managed to capture the FB pin being systematically "ringed" by inductor switching. In this picture Channel 1 is the inductor (SW pin) and 2 is the FB pin, both referenced to ground.

https://www.dropbox.com/s/uucjq7she37lz89/FB%20ringing.JPG?dl=0

Again, maybe this is expected, but the FB pin goes a full 1 volt out of regulation. Could this be the culprit?

Thnaks for your help!
doug








--
Douglas Beeson <c.difficile@gmail.com>

 

[Tim Wescott]

On Mon, 28 Sep 2015 21:33:45 -0400, Douglas Beeson wrote:

On Sun, 27 Sep 2015 14:53:19 -0500 Tim Wescott
seemywebsite@myfooter.really> wrote:

On Sat, 26 Sep 2015 20:05:08 -0400, Douglas Beeson wrote:

Hi all,

I am trying to get a TI TLV61220 step-up voltage converter to work
properly under load. This is a difficult part for me to work with as
it is very small -- SOT23-6 -- and, I suspect, finnicky about layout.
My conditions are: Input: 2 AA batteries = 1.8-3 V; Output: 5 V; load
current: 20mA

I have set up a test platform on 1-sided, 1oz etched copper which is
not exactly like the recommended layout, but I think I have avoided
the most obvious problem of output current into the feedback path.
For test purposes, I have soldered the part to a SOT23-6 -> DIP6
adapter board (also provided by TI) and soldered the pins of that
adapter to my test board. Too much induction?

Unloaded, the circuit gives me the desired 5 V (5.1 actually..). But
as soon as I load it with a single LED (6 mA), the voltage droops to
3.17 V. According to the spec sheet
https://www.dropbox.com/s/kq0092xnxlozdeu/tlv61220.pdf?dl=0 , I
should be able to provide 30-40 mA of current at 5V. I am using the
recommended cap and inductor values.

Here is the circuit:
https://www.dropbox.com/s/uu1zerdq7msfkpt/Beeson_TLV61220.pdf?dl=0

Where should I start looking for the cause and a fix?

Thank you!

doug

If you're really using an electrolytic cap -- stop. Surface-mount
ceramic 10uF caps are vastly cheap these days, and are almost certainly
what's intended for this circuit.

I have tried it with electrolytic, 10uF ceramic, and both in parallel.
No change in the output voltage sag.

It isn't that, then.

Come to think of it, though, it could be C18 -- if you haven't tried that
with a ceramic yet you may want to.

A 'lytic will give you a horrible amount of ripple at the switching
frequency, and may well be confusing the regulator into thinking that
it's holding the set point just fine.

I haven't used this exact part but I've used ones like it and not had
trouble, so I'm not sure why it's giving you fits.

I had a similar problem with another TI part that turned out to be
simnply me overloading the chip beyond its regulation ability. That one
used a lot lower switching frequency, though.

I am thinking maybe this is a problem of coupling between the switching
inductor and the feedback pin. Here's a scope capture I did with the
inductor SW pin on Channel 1 and the FB pin on Channel 2.

https://www.dropbox.com/s/uucjq7she37lz89/FB%20ringing.JPG?dl=0

Does this seem right?

It could be a layout issue. Ideally the ground ends of C18 and C19 would
be on top of the chip ground, and the hot ends of C18 and C19 would be
close to their respective pins on the chip.

If you can figure out how to monitor the coil current you may find some
joy there -- if the chip limits coil current, and if the coil is
saturating, that could cause issues like you're seeing. Or if the coil
is just plain under-valued.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

 

[Douglas Beeson]

On Sun, 27 Sep 2015 14:53:19 -0500
Tim Wescott <seemywebsite@myfooter.really> wrote:

On Sat, 26 Sep 2015 20:05:08 -0400, Douglas Beeson wrote:

Hi all,

I am trying to get a TI TLV61220 step-up voltage converter to work
properly under load. This is a difficult part for me to work with as it
is very small -- SOT23-6 -- and, I suspect, finnicky about layout. My
conditions are: Input: 2 AA batteries = 1.8-3 V; Output: 5 V; load
current: 20mA

I have set up a test platform on 1-sided, 1oz etched copper which is not
exactly like the recommended layout, but I think I have avoided the most
obvious problem of output current into the feedback path. For test
purposes, I have soldered the part to a SOT23-6 -> DIP6 adapter board
(also provided by TI) and soldered the pins of that adapter to my test
board. Too much induction?

Unloaded, the circuit gives me the desired 5 V (5.1 actually..). But as
soon as I load it with a single LED (6 mA), the voltage droops to 3.17
V. According to the spec sheet
https://www.dropbox.com/s/kq0092xnxlozdeu/tlv61220.pdf?dl=0 , I should
be able to provide 30-40 mA of current at 5V. I am using the recommended
cap and inductor values.

Here is the circuit:
https://www.dropbox.com/s/uu1zerdq7msfkpt/Beeson_TLV61220.pdf?dl=0

Where should I start looking for the cause and a fix?

Thank you!

doug

If you're really using an electrolytic cap -- stop. Surface-mount
ceramic 10uF caps are vastly cheap these days, and are almost certainly
what's intended for this circuit.

I have tried it with electrolytic, 10uF ceramic, and both in parallel. No change in the output voltage sag.

A 'lytic will give you a horrible amount of ripple at the switching
frequency, and may well be confusing the regulator into thinking that
it's holding the set point just fine.

I haven't used this exact part but I've used ones like it and not had
trouble, so I'm not sure why it's giving you fits.

I had a similar problem with another TI part that turned out to be simnply me overloading the chip beyond its regulation ability. That one used a lot lower switching frequency, though.

I am thinking maybe this is a problem of coupling between the switching inductor and the feedback pin. Here's a scope capture I did with the inductor SW pin on Channel 1 and the FB pin on Channel 2.

https://www.dropbox.com/s/uucjq7she37lz89/FB%20ringing.JPG?dl=0

Does this seem right?

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

--
Douglas Beeson <c.difficile@gmail.com>

 

[Dimitrij Klingbeil]

On 29.09.2015 03:29, Douglas Beeson wrote:
....

I also managed to capture the FB pin being systematically "ringed"
by inductor switching. In this picture Channel 1 is the inductor (SW
pin) and 2 is the FB pin, both referenced to ground.

https://www.dropbox.com/s/uucjq7she37lz89/FB%20ringing.JPG?dl=0

Again, maybe this is expected, but the FB pin goes a full 1 volt out
of regulation. Could this be the culprit?

That looks much too correlated to me - the waveform is essentially
identical aside from the initial/final DC values. That should not
happen between FB and SW given their different loads and source
impedances unless it's just common-mode ground bounce in disguise.

So I think that you are actually seeing -1 V peak of bounce on GND,
which correspondingly translates into +1 V peak pulses on all other
nodes that are referenced to GND and in your measurements.

My advice would therefore be to make the GND connection much more beefy.
Try to solder a piece of desoldering braid (if you are careful to not
saturate it with solder it's almost as good as Litz wire when it comes
to RF) between the GND pin of the IC and your common circuit GND,
and then see if the problem gets less.

Regards
Dimitrij

 

[Tim Wescott]

On Tue, 29 Sep 2015 21:51:07 +0200, Dimitrij Klingbeil wrote:

On 29.09.2015 03:29, Douglas Beeson wrote:
...
I also managed to capture the FB pin being systematically "ringed" by
inductor switching. In this picture Channel 1 is the inductor (SW
pin) and 2 is the FB pin, both referenced to ground.

https://www.dropbox.com/s/uucjq7she37lz89/FB%20ringing.JPG?dl=0

Again, maybe this is expected, but the FB pin goes a full 1 volt out of
regulation. Could this be the culprit?

That looks much too correlated to me - the waveform is essentially
identical aside from the initial/final DC values. That should not happen
between FB and SW given their different loads and source impedances
unless it's just common-mode ground bounce in disguise.

So I think that you are actually seeing -1 V peak of bounce on GND,
which correspondingly translates into +1 V peak pulses on all other
nodes that are referenced to GND and in your measurements.

My advice would therefore be to make the GND connection much more beefy.
Try to solder a piece of desoldering braid (if you are careful to not
saturate it with solder it's almost as good as Litz wire when it comes
to RF) between the GND pin of the IC and your common circuit GND,
and then see if the problem gets less.

And bear in mind that it may be a problem with the scope connection --
before you believe those traces entirely you should clip the ground lead
to the ground pin on the chip.

But -- yea. Not only are the currents high but the di/dt is high, so
ground bounce is something that must be considered.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com