Maker Pro
Maker Pro

Increasing a DC-DC conv output from 5.00 to 5.25V

P

Peter

Jan 1, 1970
0
I have a multi output power supply box, 28V DC in, and inside are
several DC-DC converters from firms like TRACO.

It is used to power various bits in a light aircraft e.g. a Lenovo T2
tablet, a satellite phone, etc.

The problem I have is that Lenovo have cheated on their USB charging.
Unlike Apple who cheated by using four resistors on the data pins
http://www.euroga.org/forums/website/714-5v-2a-usb-charging
(but who can charge from 5.0V) Lenovo need 5.25V. In fact their own
USB chargers output 5.49V (no load).

I thought about lifting the output up by 0.6V with a diode through
which the input current passes (then I have 5.6V which I can use) but
since the switch-mode DC-DC converter's input current will always be
less than the output current, the net current through the diode will
be below zero...

I cannot see any way of doing it...

Obviously with a DC-DC conv which has resistor-programmable o/p
voltage, or even remote sense terminals which can be fooled with
resistors, it's easy. But the smaller modules don't have this.

I have a "Lineage Power" open-style DC-DC conv which can do 5V-7V at
up to 20A, but it's a bit too big, has no shielding, and would need
tons of i/p and o/p filtering.

Is there some trick way of doing it?

I know how to make a 0.25V or 0.5V dropper, with a transistor... so
that bit is not a problem.
 
N

Nico Coesel

Jan 1, 1970
0
Peter said:
I have a multi output power supply box, 28V DC in, and inside are
several DC-DC converters from firms like TRACO.

It is used to power various bits in a light aircraft e.g. a Lenovo T2
tablet, a satellite phone, etc.

The problem I have is that Lenovo have cheated on their USB charging.
Unlike Apple who cheated by using four resistors on the data pins
http://www.euroga.org/forums/website/714-5v-2a-usb-charging
(but who can charge from 5.0V) Lenovo need 5.25V. In fact their own
USB chargers output 5.49V (no load).

I thought about lifting the output up by 0.6V with a diode through
which the input current passes (then I have 5.6V which I can use) but
since the switch-mode DC-DC converter's input current will always be
less than the output current, the net current through the diode will
be below zero...

I cannot see any way of doing it...

Obviously with a DC-DC conv which has resistor-programmable o/p
voltage, or even remote sense terminals which can be fooled with
resistors, it's easy. But the smaller modules don't have this.

The smaller modules output a higher voltage at low loads. Changes are
the output voltage is already suitable for your purpose.
 
L

Lasse Langwadt Christensen

Jan 1, 1970
0
Den søndag den 27. oktober 2013 20.29.00 UTC+1 skrev Peter:
I have a multi output power supply box, 28V DC in, and inside are

several DC-DC converters from firms like TRACO.



It is used to power various bits in a light aircraft e.g. a Lenovo T2

tablet, a satellite phone, etc.



The problem I have is that Lenovo have cheated on their USB charging.

Unlike Apple who cheated by using four resistors on the data pins

http://www.euroga.org/forums/website/714-5v-2a-usb-charging

(but who can charge from 5.0V) Lenovo need 5.25V. In fact their own

USB chargers output 5.49V (no load).



I thought about lifting the output up by 0.6V with a diode through

which the input current passes (then I have 5.6V which I can use) but

since the switch-mode DC-DC converter's input current will always be

less than the output current, the net current through the diode will

be below zero...



I cannot see any way of doing it...



Obviously with a DC-DC conv which has resistor-programmable o/p

voltage, or even remote sense terminals which can be fooled with

resistors, it's easy. But the smaller modules don't have this.



I have a "Lineage Power" open-style DC-DC conv which can do 5V-7V at

up to 20A, but it's a bit too big, has no shielding, and would need

tons of i/p and o/p filtering.



Is there some trick way of doing it?



I know how to make a 0.25V or 0.5V dropper, with a transistor... so

that bit is not a problem.

how would you increase the output voltage with a diode drop?

unlesss you can get to the sense resistor inside I don't see how

maybe something like this: http://www.ebay.com/itm/NEW-DC-DC-A...781?pt=LH_DefaultDomain_0&hash=item53fc77e28d

-Lasse
 
M

miso

Jan 1, 1970
0
I thought about lifting the output up by 0.6V with a diode through
which the input current passes (then I have 5.6V which I can use) but
since the switch-mode DC-DC converter's input current will always be
less than the output current, the net current through the diode will
be below zero...

In my opinion, no. BTW, I don't follow this paragraph from your original
post. But diodes are temperature sensors, so I don't like the idea, even
if I totally understood it.

To be honest, this sounds like a design flaw in the notebook. That is,
the charger in the notebook need to take 5V +/- 5% and deliver 4.2V plus
the burden of the current sense. This should be doable. I'm assuming the
Lenovo doesn't have Lion batteries in series, otherwise I would find it
hard to imagine a few tenths of a volt would matter if they are boosting
the input voltage.

My opinion of Lenovo has dropped a notch, though their customers swear
their notebooks are the best, mostly because of their keyboards. My
opinion of Apple hasn't changed. Apple loves proprietary stuff.
Basically **** the customer.
 
J

Jasen Betts

Jan 1, 1970
0
how would you increase the output voltage with a diode drop?

With a LM7805, you put it in the ground lead.
Oobviously this is not going to work with a buck
switcher.
 
D

Don Y

Jan 1, 1970
0
Hey there, Peter!

I have a multi output power supply box, 28V DC in, and inside are
several DC-DC converters from firms like TRACO.

Presumably, you want to tweek this existing box (part of the avionics
package) and not create something entirely new? E.g., a separate
"outboard" DC-DC converter (off of one of the voltages you have
available already)?

Does whatever you do have to pass any sort of certification?
It is used to power various bits in a light aircraft e.g. a Lenovo T2
tablet, a satellite phone, etc.

The problem I have is that Lenovo have cheated on their USB charging.
Unlike Apple who cheated by using four resistors on the data pins
http://www.euroga.org/forums/website/714-5v-2a-usb-charging
(but who can charge from 5.0V) Lenovo need 5.25V. In fact their own
USB chargers output 5.49V (no load).

Is it that the tablet won't sense that it *should* charge? Or,
that it *can't* charge with that low of a potential?
I thought about lifting the output up by 0.6V with a diode through
which the input current passes (then I have 5.6V which I can use) but
since the switch-mode DC-DC converter's input current will always be
less than the output current, the net current through the diode will
be below zero...

I cannot see any way of doing it...

Learn to have lunch within DRIVING distance of your home -- instead
of halfway across the Continent! ;-) ;-)
Obviously with a DC-DC conv which has resistor-programmable o/p
voltage, or even remote sense terminals which can be fooled with
resistors, it's easy. But the smaller modules don't have this.

Can you *replace* a module? Or, are there mechanical issues involved?
Can you gut a module, modify it and install that in its place?
I have a "Lineage Power" open-style DC-DC conv which can do 5V-7V at
up to 20A, but it's a bit too big, has no shielding, and would need
tons of i/p and o/p filtering.

Is there some trick way of doing it?

Off the cuff, I'd just suggest something "stand-alone" that handled
*just* the tablet -- as *it* seems to be the oddball.

Is there any other charging path into the tablet (e.g., 12VDC)
I know how to make a 0.25V or 0.5V dropper, with a transistor... so
that bit is not a problem.

If this is your biggest problem, I'll assume life is treating you
reasonably well! (living arrangement?)

Regards,
--don
 
K

Kevin McMurtrie

Jan 1, 1970
0
Peter said:
I have a multi output power supply box, 28V DC in, and inside are
several DC-DC converters from firms like TRACO.

It is used to power various bits in a light aircraft e.g. a Lenovo T2
tablet, a satellite phone, etc.

The problem I have is that Lenovo have cheated on their USB charging.
Unlike Apple who cheated by using four resistors on the data pins
http://www.euroga.org/forums/website/714-5v-2a-usb-charging
(but who can charge from 5.0V) Lenovo need 5.25V. In fact their own
USB chargers output 5.49V (no load).

I thought about lifting the output up by 0.6V with a diode through
which the input current passes (then I have 5.6V which I can use) but
since the switch-mode DC-DC converter's input current will always be
less than the output current, the net current through the diode will
be below zero...

I cannot see any way of doing it...

Obviously with a DC-DC conv which has resistor-programmable o/p
voltage, or even remote sense terminals which can be fooled with
resistors, it's easy. But the smaller modules don't have this.

I have a "Lineage Power" open-style DC-DC conv which can do 5V-7V at
up to 20A, but it's a bit too big, has no shielding, and would need
tons of i/p and o/p filtering.

Is there some trick way of doing it?

I know how to make a 0.25V or 0.5V dropper, with a transistor... so
that bit is not a problem.

I'd use one of Ti's "Simple Switcher" chips. The low voltage output
chips even have an integrated inductor so all you add is the feedback
divider and filter capacitors. Overall you'll end up with a circuit
board about the size of an SD card.

There are lots of variations but here is one to start with:
http://www.ti.com/lit/ds/symlink/lmz23603.pdf
 
P

Peter

Jan 1, 1970
0
Don Y said:
Hey there, Peter!



Presumably, you want to tweek this existing box (part of the avionics
package) and not create something entirely new? E.g., a separate
"outboard" DC-DC converter (off of one of the voltages you have
available already)?

I have a 5V 2A DC-DC conv already in there and would like to "lift"
its output by about 1/2 volt.

I don't have room in the box for one of the much bigger converters
that have an adjustable output.
Does whatever you do have to pass any sort of certification?

No - portable kit is OK. Plugs into a cigar lighter socket.
Is it that the tablet won't sense that it *should* charge? Or,
that it *can't* charge with that low of a potential?

I don't know. It draws some current from 5.0V but not enough to charge
while turned on.
Learn to have lunch within DRIVING distance of your home -- instead
of halfway across the Continent! ;-) ;-)

Hohoho :)
Can you *replace* a module? Or, are there mechanical issues involved?
Can you gut a module, modify it and install that in its place?

Not enough room.
Off the cuff, I'd just suggest something "stand-alone" that handled
*just* the tablet -- as *it* seems to be the oddball.

I am going that way, yes.
Is there any other charging path into the tablet (e.g., 12VDC)

No, sadly, because in the same box I also have these handy outputs

10V 1A
12V 1A
19V 4A
If this is your biggest problem, I'll assume life is treating you
reasonably well! (living arrangement?)

Indeed :)
 
P

Peter

Jan 1, 1970
0
miso said:
To be honest, this sounds like a design flaw in the notebook. That is,
the charger in the notebook need to take 5V +/- 5% and deliver 4.2V plus
the burden of the current sense. This should be doable. I'm assuming the
Lenovo doesn't have Lion batteries in series, otherwise I would find it
hard to imagine a few tenths of a volt would matter if they are boosting
the input voltage.

Yes - crap design.
My opinion of Lenovo has dropped a notch, though their customers swear
their notebooks are the best, mostly because of their keyboards. My
opinion of Apple hasn't changed. Apple loves proprietary stuff.
Basically **** the customer.

I still think Thinkpad laptops are the best of them all. We use them
at work and I have a few around the place myself.

I bought the Lenovo tablet because I needed a windoze platform to get
DUN networking to talk to a satellite phone over RS232 and to run some
apps which need windoze.
 
In my opinion, no. BTW, I don't follow this paragraph from your original
post. But diodes are temperature sensors, so I don't like the idea, even
if I totally understood it.

To be honest, this sounds like a design flaw in the notebook. That is,
the charger in the notebook need to take 5V +/- 5% and deliver 4.2V plus
the burden of the current sense. This should be doable. I'm assuming the
Lenovo doesn't have Lion batteries in series, otherwise I would find it
hard to imagine a few tenths of a volt would matter if they are boosting
the input voltage.

It's not clear to me the +5.5v requirement is proven. +5.0v
should be more than enough, as you noted.

Given all the different, incompatible "cheats" to enable/request
high current from the charger, it still seems possible that that's
the problem.

Cheers,
James Arthur
 
He is in an airplane, probably there is no stock charger. However, a 28V

to 12V step down plus a charger designed to work off of 12V might do the

trick.

Lenovo has a bunch of different schemes for charging their PCs and Tablets,many of them are smart and communicate with the host or receive communications from the host. The lowest tech but most reliable way of doing this is to obtain their stock charger and figure out the required DC/DC converter to power it at an alternate input.
 
"but
since the switch-mode DC-DC converter's input current will always be
less than the output current, the net current through the diode will
be below zero... "

No. That doesn't mean that. It all depends on where the reference is. However if -20-1110 or whatever, the five pin jobs, adding a diode at the bottomwill not work anyway, the reference and everything is inside.

If you can't get inside such a unit, you are up the creek and paddles are on backorder.

the current is not very high, so really you could add a diode to the outputs after a small inductor, and then throw in some AC from an oscillator or something. Umm, I guess I would recommend an oscillator over an or someting in this case. Just capacitively couple it. If you need to, post regulate, well wait, that is not what I mean. you HAVE inductance so PWN in your little gizwhicher would work fine. Damn shame to complicate it so much but that's the breaks.
 
Top