Someone clever please help boost Novatac 120P output current and keep UI intact?

Moddoo

Flashlight Enthusiast
Joined
Nov 19, 2008
Messages
1,243
Location
Minnesota
Hello guys,
I would like to put a P7 or MC-E into a 120P

Can anyone help me mod the stock driver for more current?

Thanks for any help
 

Moddoo

Flashlight Enthusiast
Joined
Nov 19, 2008
Messages
1,243
Location
Minnesota
Norm,
I enjoy your not so subtle sarcasm.
Reminds me of someone else that I am.

Perhaps the word clever brought you in here?:p

I was aware that people are happy with the emitter swap.

If you find any info on how to increase the output current on the stock driver, please share.
 

Norm

Retired Administrator
Joined
Jun 13, 2006
Messages
9,512
Location
Australia
I know the first link doesn't match your requirements exactly but it's the only thread I know of that runs the P7 with more current than the standard driver. :shrug:
Norm
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Hello guys,
I would like to put a P7 or MC-E into a 120P

Can anyone help me mod the stock driver for more current?

Thanks for any help

Send a photo of the stock driver. Make sure that all markings on the components are readable. My guess is that the light uses a boost or a boost-buck IC.

The usual approach would probably be to change out the sense resistors to give you greater output.

The challenge is for the IC to have a high enough switch current rating to run at something like 1.5A or 2A out while driven by one 1A (edit: I meant one 123A) or one Li-ion. I think it would be easier to shoot for 1A out to the multi-die LED. You'll still get 300 lm.
 
Last edited:

Moddoo

Flashlight Enthusiast
Joined
Nov 19, 2008
Messages
1,243
Location
Minnesota
I know the first link doesn't match your requirements exactly but it's the only thread I know of that runs the P7 with more current than the standard driver. :shrug:
Norm
Thanks Norm, no offense meant.

Send a photo of the stock driver. Make sure that all markings on the components are readable. My guess is that the light uses a boost or a boost-buck IC.

The usual approach would probably be to change out the sense resistors to give you greater output.

The challenge is for the IC to have a high enough switch current rating to run at something like 1.5A or 2A out while driven by one 1A or one Li-ion. I think it would be easier to shoot for 1A out to the multi-die LED. You'll still get 300 lm.

Thanks a lot man.
I'll get some good pics when the patient arrives.
Anyone have a clue what the 120P might have for output current stock?
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Anyone have a clue what the 120P might have for output current stock?

I believe that the 120P uses a Seoul P4 U flux bin LED. If the 120 lm rating is an emitter rating, then the drive current is probably around 500mA.
 

Blindasabat

Flashlight Enthusiast
Joined
Jan 24, 2006
Messages
2,204
Location
Michigan
This would be very interesting for just boosting the output to a single die emitter. I have an S2 bin (~60L @ 350mA) high CRI SSC modded NT 120 that could use a boost to get back over 100L.

BTW, I think they ran T-bin LEDs (at least at first) because many people did U-bin upgrades on the first NT120's for a noticeable increase in output. So they may be running more than 500mA in some cases (if they are tuned to 120L like HDS).
 
Last edited:

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
This would be very interesting for just boosting the output to a single die emitter. I have an S2 bin (~60L @ 350mA) high CRI SSC modded NT 120 that could use a boost to get back over 100L.

BTW, I think they ran T-bin LEDs (at least at first) because many people did U-bin upgrades on the first NT120's for a noticeable increase in output. So they may be running more than 500mA in some cases (if they are tuned to 120L like HDS).

I searched CPF on this issue and the results were that some saw an output increase, others did not. It appeared to be a flux lottery issue as to what result one got. The U-bin is quite wide -- 91-118.5 lumens at 350mA drive current. One poster also contacted Novatac, and they responded that they used U-bin. This was only a month after release, so some speculated that it seemed hard to believe that they'd switch flux bins (and presumably alter the driver to stay at 120 lumens) after such a short period on the market.

Let's see if 500mA drive current is consistent with the run time in high mode.

efficiency * Vbatt * Ibatt = Vf * If

Let's assume

- 80% driver efficiency
- I Vf bin P4
- If = 500mA; thus Vf~3.4V
- Vbatt (1x123A) = 2.5V

Thus, Ibatt = (3.4*0.5)/(0.8*2.5) = 0.85A

Looking at the various CR123A curves in Silverfox's 123 shootout, I can believe that the Novatac falls out of regulation after about 30 min.

A T-bin would need about 750mA drive current to reach roughly 120 emitter lumens. That would give a value of Ibatt~1.3A, which seems less consistent with the claimed run times for the 120P.

Of course, this is all guesswork.
 
Last edited:

Moddoo

Flashlight Enthusiast
Joined
Nov 19, 2008
Messages
1,243
Location
Minnesota
ah crap, don't make me go tear this apart and take a drive current reading!!!

Really though, I came up with a similar conclusion JC.

Nobody seems to know.

I'll dissect the new patient this weekend and take a reading or 2.
However imperfect a simple DMM measurement might be.
It will be close enough for me.

Not opening up my EDC again right now.:eek:oo:
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Maybe you can take a tail current reading first. A U bin vs T bin seem sufficiently different that a tail current measurement might distinguish between the two. If you do conduct some surgery on the light, don't forget to take a photo of the driver.
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
This post measured 690mA current draw.

Let's try to bracket some variables. At 690mA, let's assume Vbatt=2.6V.

First, let's look at the estimated 500mA case. Suppose the P4 was H voltage bin, not I bin:

Ibatt ~ (3.2*0.5)/(0.8*2.6) = 0.77A

Still higher than the measured data.

Let's improve the driver efficiency to 90%:

Ibatt = 0.68A

Now, let's look at the estimated 750mA case.

Ibatt ~ (3.2*0.75)/(0.9*2.6) = 1A (and actually, Vbatt would be closer to 2.5V again at this discharge level)

My guess is that the LED is U bin and that the drive current is around 500mA.
 

tebore

Flashlight Enthusiast
Joined
May 10, 2006
Messages
2,141
Location
Toronto, Ontario. CAN.
You guys should know it's very difficult to remove the electronics while keeping it intact.

I remember that post I believe it was done with an RCR so you're gonna have to bump the voltage up to 3.6 underload. I did some testing a long time ago and the output to the emitter is close to 700mA. The older HDS EDC pumped close to 1A to the emitter. At 700mA using the old U bin you're around 120-140lm at the emitter. I remember reading that when the first NT's came out they were actually shy of the 120lm rating ~110lm.

There's really no point to upping the current to the LED, the light doesn't have that much mass to heatsink it and the efficiency really suffers. The Efficiency of the NT driver is already less than stellar. If you're looking for a 10-20% boost in output and a tighter hotspot just put in a K2 TFFC.
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Good to see a figure based on actual measurements, rather than guesswork. But 700mA to a U bin P4 puts the relative luminous flux at about 1.8. The U flux bin is 91-118.5 lm at 350mA. Multiplying by 1.8 gives about 160-210 emitter lumens. So maybe the Novatac 120 lumen figure is an OTF rating, not emitter lumens (I had assumed emitter lumens, leading me to believe that the driver current was closer to 500mA)? A T flux bin P4 would give about 120 emitter lumens at 700mA drive current, which someone else had suggested before and which I had discounted.
 

Russki

Enlightened
Joined
Oct 18, 2006
Messages
394
Location
New York
You guys should know it's very difficult to remove the electronics while keeping it intact.

I remember that post I believe it was done with an RCR so you're gonna have to bump the voltage up to 3.6 underload. I did some testing a long time ago and the output to the emitter is close to 700mA. The older HDS EDC pumped close to 1A to the emitter. At 700mA using the old U bin you're around 120-140lm at the emitter. I remember reading that when the first NT's came out they were actually shy of the 120lm rating ~110lm.

There's really no point to upping the current to the LED, the light doesn't have that much mass to heatsink it and the efficiency really suffers. The Efficiency of the NT driver is already less than stellar. If you're looking for a 10-20% boost in output and a tighter hotspot just put in a K2 TFFC.

Thanks tebore :wave:
Why I everybody hooked up on quadrisect emitters?
K2 TFFC id definitely way to go, for HDS lights.
But I am still enjoying my Cree R2 HDS 42 mod from 2007.
:D
 

tebore

Flashlight Enthusiast
Joined
May 10, 2006
Messages
2,141
Location
Toronto, Ontario. CAN.
For the SSC you should always take the lowest value of the bin. The U-bin was good for about 90lm at 350mA at 1.8x at 700ma it's around 160lm. The thing is the SSC specs are very very loose. The real value at 700mA with the first u-bins was very more like 1.7x the output at 350mA at best. I should look at JTR's testing to see what the new U2 and U1 bins can do.

The NT's 120Lm rating is pretty much an OTF approximation. Testing shows 100lm - 120lm on most.
 

Moddoo

Flashlight Enthusiast
Joined
Nov 19, 2008
Messages
1,243
Location
Minnesota
Good stuff guys.

I carry my 120P with a K2 tffc everyday.

It's great.

I now have another 120P that was purchased specifically to mod.

I have the LEDs waiting.
 

DM51

Flashaholic
Joined
Oct 31, 2006
Messages
13,338
Location
Borg cube #51
There will be occasions when that extra output is very welcome and useful, but for most situations it is not required - and in fact it isn't even very easy to tell the difference.

I did a test last night with 3 family members using an HDS B42, HDS U60 and Novatac 120. When all 3 were on at the same time, it was not too difficult to tell which was which, due to the relative brightness, although even then one of my children thought the B42 was brighter than the U60.

When they were turned on in random sequence, all 3 family members made at least one mistake in ranking them in order of brightness. One family member even put the HDS-42 ahead of the NT-120 (although on repeating the same test twice, this didn't happen again).

I'm not trying to prove anything, as that was a very unscientific test, but it is my observation and experience that those extra lumens don't make a lot of difference unless there is a pretty substantial increase - probably double or more - and the tint and beam pattern will have a big influence on individual perception of brightness.
 
Top