new P7 driver 3amps?

CM

Flashlight Enthusiast
Joined
Sep 11, 2002
Messages
3,454
Location
Mesa, AZ
Its is to good to be true. :mecry:
But 83,5% is good enough for me :grin2:

I have measured the thing and I have to say I like it.
Unlike I have read somewhere else these drivers do have a memory function.
They will start working from 4,7Volts, they will stop working at 4,0Volts than the driver goes into a verry verry low mode (3mA to the led).
I did'nt see any flickering :D

Well here are the measurements in a excel file.

All measurements where made with calibrated meters.

Greetz Johan

If you look at the amount of power lost due to efficiency (or lack of it) if you look at the high setting, this is what I get from your spreadsheet:

Efficiency Power loss
83.5% 2.1152
83.0% 2.0808
82.4% 2.1282
83.5% 2.008
81.3% 2.2855
80.8% 2.3628
79.8% 2.4992
79.0% 2.6214
78.7% 2.6901
77.8% 2.8201
76.6% 3.0174

At best efficiency of 83.5%, 2.1W of power is a lot of heat.

For flashlight use, I think that the thinking has to change for the P7. I'm using a driver with ~92% efficiency and the heat generated due to inefficiency is 33% of that compared to the cheap driver with 83% efficiency. The converter gets very warm dissipating about 830 mW of power but it can be run without a heat sink safely. On the other hand, 2400 mW with the inefficient driver means an early demise for the driver. I think a driver with efficiencies in the low 80's driving a 12W load is a poor design but probably good enough for the masses.
 

mitch79

Newly Enlightened
Joined
Jul 11, 2008
Messages
134
Location
Adelaide, South Australia
I agree CW. The problem is I have yet to find any other P7 driver that will fit in a "C" Mag powered by 2x Li-Ion.

I'm using one of Der Wichtel's drivers in my "D" mag but it won't physically fit in the smaller "C" tube.
If there's a better driver out there, the someone tell me please.
 

Der Wichtel

Enlightened
Joined
Mar 12, 2007
Messages
829
Location
Germany
And what about putting the driver into the batterytube?

Then the driver have to be stablized with epoxy or something similar to handle the pressure made by the tailcapspring

I don't have a C Mag ( only a very old one with wider diameter), so it's just an idea.
 
Last edited:

Aircraft800

Flashlight Enthusiast
Joined
Apr 24, 2007
Messages
1,487
Location
DFW Texas.
I agree CW. The problem is I have yet to find any other P7 driver that will fit in a "C" Mag powered by 2x Li-Ion.

I'm using one of Der Wichtel's drivers in my "D" mag but it won't physically fit in the smaller "C" tube.
If there's a better driver out there, the someone tell me please.

My problem exactly, I love the Small Form Factor of my 2C M*g, but the cheap 17mm buck driver that I have shuts down after 5min. use.

I need to try to determine which components are getting hot, and try to heatsink them better, or someone needs to release a more efficient buck driver for the P7. The Der Wichtel one looks like the best one out, but I haven't seen anyone able to fit it between the H22A P7 heatsink and the M*g switch.
 

mitch79

Newly Enlightened
Joined
Jul 11, 2008
Messages
134
Location
Adelaide, South Australia
Looking at Aircraft800's previous post, the area marked as getting hot looks like a pair of diodes. Now why would a pair of diodes be getting hot?

Methinks the full 2.8A current is passing through them and their being pushed right to the edge of their current handling ability.

The other question is what are they being used for? Is it just reverse polarity protection? If so I'd be happy to remove them.

Anyone here with more electronics experience than me who can nut out this circuit and tell us what's happening?

EDIT: Googling "SS34" indicates it's a 3A 40V Schottky diode. Hope that helps.

Cheers,

Mitch.
 
Last edited:

Aircraft800

Flashlight Enthusiast
Joined
Apr 24, 2007
Messages
1,487
Location
DFW Texas.
Looking at Aircraft800's previous post, the area marked as getting hot looks like a pair of diodes. Now why would a pair of diodes be getting hot?

Methinks the full 2.8A current is passing through them and their being pushed right to the edge of their current handling ability.

The other question is what are they being used for? Is it just reverse polarity protection? If so I'd be happy to remove them.

Anyone here with more electronics experience than me who can nut out this circuit and tell us what's happening?

EDIT: Googling "SS34" indicates it's a 3A 40V Schottky diode. Hope that helps.

Cheers,

Mitch.

Mitch,

You know what this means? I heatsinked the wrong components! :mecry:
I used a picture from another post to determine where to put it, no testing. Those diodes probably don't get hot at all, but I was anxious to assemble my light.

Before I disassemble my light, do you have a clew where the heat is generated? The IC? Which is it from those pictures?

Thanks for your help guys!!
 

Tohuwabohu

Newly Enlightened
Joined
May 8, 2008
Messages
184
Location
Germany
Looking at Aircraft800's previous post, the area marked as getting hot looks like a pair of diodes. Now why would a pair of diodes be getting hot?

Methinks the full 2.8A current is passing through them and their being pushed right to the edge of their current handling ability.

The other question is what are they being used for? Is it just reverse polarity protection? If so I'd be happy to remove them.

Anyone here with more electronics experience than me who can nut out this circuit and tell us what's happening?

EDIT: Googling "SS34" indicates it's a 3A 40V Schottky diode. Hope that helps.

Cheers,

Mitch.

The Schottky diode is an essential part of the buck converter.
At 3A the SS34 has a forward voltage of approximately 0.4V.
Approximately 0.4V * 2.5A = 1W of power are dissipated in a very small component, that's why it is getting so hot.
Putting 2 diods in parallel directly side by side doesn't make ich very much better.

Aircraft800,
I think you did heatsink the right components.
 
Last edited:

Northern Lights

Flashlight Enthusiast
Joined
Jan 17, 2006
Messages
1,267
Location
Southwest
Really? $12 is expensive in comparison to the the rest of the light? In my opinion, ALL of these driver boards I've found are a steal! Even the best D2DIM is only $20.

Let's say you were going to make a Mag 3D P7 light:
$20 Mag 3D
$27 P7
$20 heat sink
$12 driver
$36 batteries
$15 MOP reflector
$6 lens
$5 charging jack
$2 thermal adhesive
$3 bicycle tubing
$5 wire
$20 shipping
Total = $171
This $12 driver is only 7% of the overall parts cost, not to mention your time in building it.

I could not resist a reply, are you ever so correct! :broke: And you can get good results, almost same discharge graph, with high capacity DD models too. I just built a P7 with a jack,
1.5 D P7, Modes, Charging Jack, Electronic GID & more ,
and three more are coming that will sell for $225 and more. I own a P7 MTE I made DD with a five minute mod and it is almost as good in total output, very close, cost was < $50.

From a utilitarian point of view, the custom light gets a little more light out and runs longer. :shrug:
 

mitch79

Newly Enlightened
Joined
Jul 11, 2008
Messages
134
Location
Adelaide, South Australia
Thanks for the explanation Tohuwabohu :).

Having re-read this thread again I no longer believe the diodes are the problem.
1. Aircraft800 heatsinked the diodes and it still shuts down after 5 minutes.
2. A diode can't do that, it's just a semiconductor.

The only thing I can think of that could cause that effect is a programmable IC with built in temperature monitoring.
Anyone care to guess which one's overheating?
 

kavvika

Enlightened
Joined
Jan 30, 2007
Messages
509
Location
Chicago
Few questions before I purchase this driver and begin my first P7 mag-mod:

-If I place an order today, will I receive a driver with or without memory mode?
If the answer is that all new drivers are shipped with memory mode, how would I acquire a driver without memory mode?
-Also, how should I solder the LED and power wires to bypass memory mode? I know it can be done on a few other drivers I have, such as DX #7880. Can the MCU be bypassed on this drive board?
 

mitch79

Newly Enlightened
Joined
Jul 11, 2008
Messages
134
Location
Adelaide, South Australia
I received my KD P7 driver a couple of weeks ago. It's the newer 17mm version.
Like others here it overheats and cuts out to low mode after 3 minutes running on high. Power is 2x 18650 Li-Ion.

The diodes where getting way to hot, so tonight I changed them.
I used 2x 1N5822 40V 3A Schottky Diodes in parallel.

In Australia you can get them from DSE or Jaycar.

These axial lead diodes are physically much larger then the originals and will not fit on the PCB but it's not hard to solder them togeather and run a pair of short wires to them.

The result? I ran the light for 30min continuous on high before I turned it off. No problems.
Finally my 2C P7 Mag is a sucess. :)

Sorry, no photos yet. I was in a hurry to assemble the light and I wasn't sure it would work in the first place.
I can now assure you it does :twothumbs
 

wquiles

Flashaholic
Joined
Jan 10, 2005
Messages
8,459
Location
Texas, USA, Earth
If you look at the amount of power lost due to efficiency (or lack of it) if you look at the high setting, this is what I get from your spreadsheet:

Efficiency Power loss
83.5% 2.1152
83.0% 2.0808
82.4% 2.1282
83.5% 2.008
81.3% 2.2855
80.8% 2.3628
79.8% 2.4992
79.0% 2.6214
78.7% 2.6901
77.8% 2.8201
76.6% 3.0174

At best efficiency of 83.5%, 2.1W of power is a lot of heat.

For flashlight use, I think that the thinking has to change for the P7. I'm using a driver with ~92% efficiency and the heat generated due to inefficiency is 33% of that compared to the cheap driver with 83% efficiency. The converter gets very warm dissipating about 830 mW of power but it can be run without a heat sink safely. On the other hand, 2400 mW with the inefficient driver means an early demise for the driver. I think a driver with efficiencies in the low 80's driving a 12W load is a poor design but probably good enough for the masses.

+1. Glad I am not the only one to note that in terms of efficiency and wasted power, most all of these "new" drivers are simply terrible. Of course, the P7 is just bringing the worst out, given the 3Amp goal.

For now, I "still" like the D2DIM: although it is not a "regulated" driver, when using it with good LiIon cells for a nearly flat discharge, it gives me full dimming control at nearly 100% efficiency by using PWM.

Will
 

kevinm

Enlightened
Joined
Jun 2, 2006
Messages
891
Location
Denverish
So the new one is 17mm...what's the new thickness from the board to the top of the inductor and total thickness? I'm looking for something 6mm or so.

Thanks,
Kevin
 

Packhorse

Flashlight Enthusiast
Joined
Nov 29, 2007
Messages
1,912
Location
New Zealand
I have 2 of these drivers in use.
The 1st is priving a single P7 from a 6 cell NiMh pack and works great

The second is driving 4 Q5. 2P2S 1400ma each driven off a 8 NiCd pack.
This one causes a lot of interference on my TV and my landlords radio. More so when on the high setting.
Any ideas on how to fix this (apart from replacement). Will a cap on the output help?
 

uhg_

Newly Enlightened
Joined
Oct 9, 2008
Messages
14
FYI: Tried this driver with 2 resp. 3 P7s in series, using a slightly tired 12V 4A NiHm. 2S driven at the full 2.8A on the LED side; 3S driven at 2.0A with battery Vin dropping to just above 10V. Driver gets warm enough to need cooling but not more, I think, than a bit of alu and/or some potting can keep in check.

Haven't done any extended testing so proceed at your own risk. But this seems like a pretty OK driver, at least for my purposes.
 
Top