EverCool Variable Voltage Mod

[bigblue30]

Anyone that has used any of the regulator chips (fixed or variable) know how “hot” they can get. I have cut vents in the case and added heat sinks just to cool them down. All that heat is wasted power that has to come from the battery, shortening the run time of your mod.

I went looking for a chip that did not convert the extra voltage into heat and found the TI PTR08100W “http://focus.ti.com/lit/ds/symlink/ptr08100w.pdf”

It is a SWITCHING REGULATOR

Because it is a switching regulator, it does not burn off the excess voltage as heat like the UCC283-ADJ, the UCC283-5, and many other regulators that we use,so it does not get at all hot (or even warm) and it does not require a heat sink.

Some people will not like this chip because the max voltage is only 5.5 volts (6 volts if you go outside of the specs). That is OK with me because my sweet spot is 4.3 volts. Its efficiency is up to 96%, so my battery will last longer.

Here is the mod that I made using the PTR08100W chip. A lot of people have talked about this chip, but I have never seen a mod made from one -- until now!


Some of the chips specs are:

Up to 10-A Output Current
Wide Input Voltage Range (4.5 V to 14 V)
Wide-Output Voltage Adjustable (0.6 V to 5.5 V)
Efficiencies Up To 96%
ON/OFF Inhibit
Undervoltage Lockout (UVLO)
Output Overcurrent Protection
(Nonlatching, Auto-Reset)
Overtemperature Protection

Here are a few pictures and the circuit I used:

This is how it all starts on the bread board:

Here are the parts for the PCB

Chip and the PCB:

All

ready to put the PCB in the box:

Box, switch, and LED with resistor:

Rewired box with 20 AWG wire:

All

done.

 

[Rockalot]

Very very nice!

 

[Nuck]

Huge step up from using a linear regulator. It's amazing that more people haven't used this chip until now.

A set of tenergy LiPO CR123a batteries fresh of the charger only put out 5.3-5.4v under load. The 5.5v limit is more than anyone should ever need.

 

[Scubabatdan]

Sweet, how is it working? little to no heat? and is the little atty fireing good?
Thanks,
Dan

 

[roadrash]

Real mod Big Blue, I have been waiting for some one to post a how to on the efficient ptr 080s. The 5 volt boosters work ok, but they eat up bats if you are a heavy vaporer. This unit should last a long time without changing bats and no more hot spots.

 

[bigblue30]

Sweet, how is it working? little to no heat? and is the little atty fireing good?
Thanks,
Dan

Dan,

I just came in to change the battery....It has been running on this set for 30 hours.....I was only getting 20 hours on the other TI chip.

There is NO heat at all. This thing is ice cold all the time.

I am not sure what you are asking about the atty... I have it set to the same voltage as the ucc383 and it is exactly the same hit, vapor, and feel.

 

[BuzzKill]

Very cool, too bad they are $16 a piece ! or $8 @ 1,000 each

But as Nuck said NICE !

 

[bigblue30]

Very cool, too bad they are $16 a piece ! or $8 @ 1,000 each

But as Nuck said NICE !

If I had your $$$$$$ I would burn mine....Thanks

 

[Rocketman]

Blue,
That is a really nice implementation of a switching regulator. It is bound to be more efficient (when set at reasonable voltages) than a linear regulator.

The up to 96% efficiency value quoted by the manufacturer is over a rather narrow operating range, and represents the chip only. Have you attempted to measure it at your normal operating conditions? It would be interesting to know actual input and output parameters for your circuit and a linear regulator circuit (using the same test setup) to see how much energy is saved. If yours is 93% at your normal operating condition, and additional losses are 3% or less, then your 90% efficient design wastes a lot less than a linear regulator mod with 65 to 70% efficiency (for those that survive the heat ).

again, nice design, and nice implementation.

 

[Scubabatdan]

If I had your $$$$$$ I would burn mine....Thanks

Well you can order 1 as a free sample so mine is on the way.
Dan

 

[asdaq]

Nice chip, done into a tight little package. I can't quite see in the photo, but is it the 1st pin indeed unused. I would really like the efficiency along with adjustablity in a 4pin setup. Looks to be some LDO's over on the sideline of your breadboard just watching and waiting... poor things

 

[SurvivorMcGyver]

I have to agree that switching regulators are, by far, the better of all choices out there for mods.

Jesh -- they could have at least included the "required" caps on the chip {chuckling}.

And as you say -- MUCH more efficient than any of the LDO's (by themselves). More stable and consitant as well.

 

[bigblue30]

I can't quite see in the photo, but is it the 1st pin indeed unused.

Yes ....pin 1 is the "Inhibit"...If left unconnected the chip turns on whenever the input is above 4.5 volts...If you want to use it, you ground it to turn it off and open it to turn it on.

I would really like the efficiency along with adjustablity in a 4pin setup. Looks to be some LDO's over on the sideline of your breadboard just watching and waiting... poor things

Is is a TI booster chip...PTN04050C.,a LOD and a few MOSFETS for a touch switch.

 

[bigblue30]

The up to 96% efficiency value quoted by the manufacturer is over a rather narrow operating range, and represents the chip only.

That is true.

Have you attempted to measure it at your normal operating conditions? It would be interesting to know actual input and output parameters for your circuit and a linear regulator circuit (using the same test setup) to see how much energy is saved. If yours is 93% at your normal operating condition, and additional losses are 3% or less, then your 90% efficient design wastes a lot less than a linear regulator mod with 65 to 70% efficiency (for those that survive the heat ).

When I had it on the bread board with a 2.2 ohm resistor connected to the output the numbers I was getting with it set to 4.5 volts were:

input voltage, 7.6 volts and 1.3 amps of current..that = 9.88 watts of input power.

The output was 4.5 volts with 1.9 amps...that = 8.55 watts.

I need to setup a LDO and do the same test....something tells me that the numbers will not be as good.

again, nice design, and nice implementation.[/QUOTE]

 

[Rocketman]

That is true.




When I had it on the bread board with a 2.2 ohm resistor connected to the output the numbers I was getting with it set to 4.5 volts were:

input voltage, 7.6 volts and 1.3 amps of current..that = 9.88 watts of input power.
The output was 4.5 volts with 1.9 amps...that = 8.55 watts.
I need to setup a LDO and do the same test....something tells me that the numbers will not be as good.

again, nice design, and nice implementation.

[/QUOTE]

You would probably pick up a couple percent at a little higher output, plus to keep it apples vs apples match the conditions with the LDO.
This is the type of testing that provides believable results.

Comparing your results with those of someone else using different components, test equipment, even holding their mouth differently would introduce variables and potential errors.

Looking forward to seeing your comparison (when you have the time).


again, nice design, and nice implementation.

 

[bigblue30]

Looking forward to seeing your comparison (when you have the time).

I will start off by saying that I would love for someone else to bread board both of these chips and see if they get the same kind of numbers… Having said that, here is what I found…

The setup I used was this:
1- TI-PTR08100W
1- TI- UCC-383-adj …I also tested a UCC282-adj. The numbers were the same for both chips.

Both chips were set-up on the same bread board….in the same 5 holes.

Both were powered from the same 3AA box with 2 14500 900ma batteries.

Both had the caps that TI recommends.

Both tests were done with fresh batteries just off the charger. Both sets of batteries were within .001 volts of each other with no load on them.

The load for each test was a 220 ohm resistor attached to a LED to ground and a 2.2 ohm 10 watt power resistor connected from output to ground.

All readings were take with a Amprobe 34XR-A meter.

Here are the numbers:

TI-PTR08100W chip:

Input voltage 7.60v
Input amps 1.30
Input watts 9.88

Output voltage 4.5
Output amps 1.90
Output watts 8.55

Efficiency 86%

TI-UCC-383-adj:

Input voltage 7.24
Input amp 2.05
Input watts 14.84

Output voltage 4.5
Output amps 1.90
Output watts 8.55

Efficiency 57%

I believe the difference in the input voltage was because the 383 was “hitting” the batteries harder. Both sets of batteries acted the same way…when the switch was thrown the 383 always had a lower input voltage.

When I began this comparison, I expected to see a difference in efficiency, with the PTR08100W being the most efficient. I was surprised, however, to see just how much more efficient it was.

Again, if anyone thinks that these numbers might change with a different adjusted output voltage, I can rerun these tests with that voltage.

 

[WillyB]

Very sweet mod. Those efficiency numbers are amazing. The heat never bothered me, till now.

 

[bigblue30]

Very sweet mod. Those efficiency numbers are amazing. The heat never bothered me, till now.

Thanks Willy...I had to rerun the numbers a few time...I could not believe it the first time.

 

[roadrash]

Big Blue, I hope you stopped to have a beer or two whilst doing all that testing.

 

[bigblue30]

Big Blue, I hope you stopped to have a beer or two whilst doing all that testing.

That is my favorite saying..."Hold my beer....watch this"