Feedback/impressions on the PhD-M6 custom battery pack

[donn_]

I guess I didn't realize we could order the device with our own preferred target voltages programmed into the four levels.

Time for some research.

 

[wquiles]

Will, when you get the instruction document updated the way you want it, can you please post it as link to download in the sales thread? I want to replace my earlier one (or email me a copy)--no hurry. Thank you sir!

I just updated the document - available from the main sales thread. Comments/suggestions welcomed

Been thinking about the levels some more. Can you list the bulbs that you envision for Level 3, and their "optimum" (or acceptable) voltages vs the actual Level 3 voltage of 7.5V? Basically, if there is some spread of the optimum/acceptable voltages for say an MN15, MN16, MN20, and other bulbs about 7.5V, then I might want to specify re-programming of the Level 3 voltage that might more closely match some bulb of interest whose optimum/acceptable voltage might be closer to 7.1V-7.3V. If such a case exists, then this might also better accommodate a WA1111/64250 than the current Level 3 voltage of 7.5V, providing a level that can drive the bulb a little harder than Level 2 but not so hard as to risk erratic filament life.

I just talked to Eric about various things this morning and I told him about this good question that you have. In fact Eric still "owes" us a chart with a few more recommendations, so your question is very timely.

Eric said these recommendations still needed some tweaking but that based on his testing, this was a good starting point:
MN15: Range: 7.2-8.0V, recommended drive level: 7.5V
MN16: Range: 6.5-7.0V, recommended drive level: 6.8V
MN20: Range: 7.2-8.0V, recommended drive level: 7.5V
MN21: Range: 6.5-7.0V, recommended drive level: 6.8V
HO-M3T: Range: 7.2-7.9V, recommended drive level: 7.5V
EO-M3T: Range: 7.1-7.8V, recommended drive level: 7.5V
IMR-M3T: Range: 7.0-7.8V, recommended drive level: 7.5V
WA1111/64250: Range: 6.4-7.2V, recommended drive level: 6.8V

(once Eric reviews/finalizes this list, I will update the sales thread with it)

Eric said he will come back and review these once he is done running some errands this morning

I guess I didn't realize we could order the device with our own preferred target voltages programmed into the four levels.

Yes. It is an extra charge, but I can program the pack to different default voltages.

Here I am quoting myself:

Unless I hear otherwise, I am going to assume the default values/behavior for those orders:

Level 1 = 5.0 volts
Level 2 = 6.8 volts
Level 3 = 7.5 volts
Level 4 = 10.8 volts

Default behavior for almost depleted cells = Moon Mode

Moon Mode means = PWM duty cycle goes down by 1/2, so output RMS voltage goes down approximately 30%.

NOTE: Do NOT send payment until instructed by me, either here in this thread or by email (please, no PM's nor visitor messages). I will not be collecting money until I have a pack ready to ship to you.

You are nearly at the top of the list, so I need to know ASAP if you want the default values, or if you want me to bump you down until you are ready.

 

[donn_]

Yes. It is an extra charge, but I can program the pack to different default voltages.

You are nearly at the top of the list, so I need to know ASAP if you want the default values, or if you want me to bumb you down until you are ready.

Don't "bumb" me! I'll give you my target voltages shortly.

 

[wquiles]

Don't "bumb" me!

Opps, sorry for the typo. Meant bump you down

I will wait for your custom values.

 

[donn_]

Here they are, Will:

Level 1: 7.2V
Level 2: 8.5V
Level 3: 9.6V
Level 4: 10.8V

 

[wquiles]

Here they are, Will:

Level 1: 7.2V
Level 2: 8.5V
Level 3: 9.6V
Level 4: 10.8V

I updated your order - thanks for the quick turn-around

 

[mdocod]

Hi Justin Case,

Great questions. I will be working on a chart that Will can edit into the first post of either this thread, or the sales thread over the weekend. I'll come up with an estimated range of drive levels for each bulb for folks to consider.

We don't know exact bulb specifications on SF bulbs, but we can deduce a lot from looking at current draw and comparing that to discharge graphs of CR123s and working backwards. The only way to really find out exactly where bulb drive levels life ratings line up would be through destructive testing of hundreds of bulbs to find the average failure point.

the 7.5V level 3 is optimized as a conservative but impressive drive level for SF standard output lamps. (MN15, MN20). 2.5V per CR123 that the bulb is designed to run from provides what I would call an optimized drive level for these bulbs. (That's why we chose 5.0V for the N1).

Under the load of SF high output lamps, CR123s sag to something closer to 2.25V per cell in the stack. So the 6.8V setting is what I consider a nice optimum drive level for the MN16 and MN21.

Obviously everyone has preferences for balancing bulb life and efficiency. I suspect that most people, once they have tried these default settings, will agree that they are well chosen for running SF bulbs.

As it would turn out, the 7.5V setting also works great for all of the LumensFactory M3T lamps. Including the IMR-M3T. I've actually done a lot of my run-time tests on the EO-M3T, and IMR-M3T, and the bulbs are holding up well at this voltage and providing great performance.


For your specific case, I would look at the 5V and 10.8V options and decide if you can afford to "drop" one or both of them, Use that extra setting/s to sneak in an optimized drive for your favorite 1111, while retaining proper drive levels for the stock SF lamps. For example: 6.8V, 7.1V, 7.5V, 10.8V. Just throwing out ideas to ponder


Eric

 

[mdocod]

Here they are, Will:

Level 1: 7.2V
Level 2: 8.5V
Level 3: 9.6V
Level 4: 10.8V

Very curious which bulbs you have in mind. Would you mind sharing?

Eric

 

[donn_]

Very curious which bulbs you have in mind. Would you mind sharing?

Eric

Level 1: 7.2V 1111 789L/112F/11.2Hr .......5761 1237L/152F/11.2Hr

Level 2: 8.5V JC5607 1082L/111F/31Hr

Level 3: 9.6V 1164 1361L/132F/8.9Hr

Level 4: 10.8V 1185 1171L/110F/12.2Hr

 

[Justin Case]

We don't know exact bulb specifications on SF bulbs, but we can deduce a lot from looking at current draw and comparing that to discharge graphs of CR123s and working backwards. The only way to really find out exactly where bulb drive levels life ratings line up would be through destructive testing of hundreds of bulbs to find the average failure point.

You can measure tail draw with a DMM and then use a bench power supply to estimate the corresponding voltage that gives you that current draw. Of course, there can be certain complications from system resistances resulting in differing amounts of voltage drop for the flashlight vs bench supply setup.

 

[mdocod]

You can measure tail draw with a DMM and then use a bench power supply to determine the corresponding voltage that gives you that current draw.

Yes, but this information does not produce bulb life estimates, which is the primary factor in determining "optimized" drive levels.

Eric

 

[donn_]

I wouldn't say bulb life is the primary factor in determining drive levels. It's one of them, but it's no more important than unit price, output level and beam quality.

For example, the JC5607 can provide 1000+ lumens at 8.5V and live for 31 hours. But, if you want to live it up, you can take it to over 2100 lumens on 10.8V for less than 2 hours. At $1.15 each, you can afford to be a sport.:devil:

 

[mdocod]

Level 1: 7.2V 1111 789L/112F/11.2Hr .......5761 1237L/152F/11.2Hr

Level 2: 8.5V JC5607 1082L/111F/31Hr

Level 3: 9.6V 1164 1361L/132F/8.9Hr

Level 4: 10.8V 1185 1171L/110F/12.2Hr

Hi Donn,

Keep in mind, the 10.8V setting on the PhD-M6 provides soft start, but doesn't really provide meaningful regulation at the high loads presented by the 1185 and similar bulbs. Depending on the temperature and condition of the cells, the regulated run can be anywhere from non-existent, to 5-10 minutes worth.

You're ready to remove the bezel of the turbohead in order to install the 5761? They don't usually fit through the hole. Just making sure Also, at 7.2V, that's about 40W, which is pushing the cells even harder than the MN21 at ~35W, I think that's a bit too much for safety, unless it's not used much or only used on very short bursts.

The 5607 at 9.6V: That's a 3-4C load on the cells which is very unsafe. I can't even guarantee that the PhD-M6 will be able to turn that bulb on. The protection in the cells will likely prevent it.

I think Will and I are going to have to discuss putting some kind of warning or disclaimer or something in place that says we are not going to configure the PhD-M6 in scenarios where the intention is to exceed 35W power consumption.

Eric

PS: when IMR17670s come out, I don't see any problem with powering up ~50W bulbs.

 

[Justin Case]

Yes, but this information does not produce bulb life estimates, which is the primary factor in determining "optimized" drive levels.

Eric

Since we are talking about SF bulbs, it is a safe assumption that SF has done that work already. Thus, estimate the voltage at which the SF bulbs run and use that in the PhD-M6. QED.

 

[mdocod]

Since we are talking about SF bulbs, it is a safe assumption that SF has done that work already. Thus, estimate the voltage at which the SF bulbs run and use that in the PhD-M6. QED.

Yes, that's what I did, which I already said.

 

[Justin Case]

Hi Donn,

Keep in mind, the 10.8V setting on the PhD-M6 provides soft start, but doesn't really provide meaningful regulation at the high loads presented by the 1185 and similar bulbs. Depending on the temperature and condition of the cells, the regulated run can be anywhere from non-existent, to 5-10 minutes worth.

You're ready to remove the bezel of the turbohead in order to install the 5761? They don't usually fit through the hole. Just making sure Also, at 7.2V, that's about 40W, which is pushing the cells even harder than the MN21 at ~35W, I think that's a bit too much for safety, unless it's not used much or only used on very short bursts.

The 5607 at 9.6V: That's a 3-4C load on the cells which is very unsafe. I can't even guarantee that the PhD-M6 will be able to turn that bulb on. The protection in the cells will likely prevent it.

I think Will and I are going to have to discuss putting some kind of warning or disclaimer or something in place that says we are not going to configure the PhD-M6 in scenarios where the intention is to exceed 35W power consumption.

Eric

PS: when IMR17670s come out, I don't see any problem with powering up ~50W bulbs.

All of my Philips 5761 bulbs fit in my KT4. As do all of my 64250s. Just lucky I guess.

As your "PS" suggests, I assumed that Donn was planning ahead for the day that IMR17670 cells become available. Then his PhD-M6 would be already setup and optimized for high power use.

1164 will probably draw over 4A at 9.6V.

5761 will draw over 5A at ~7V.

1111/64250 will draw almost 4A at the low 7V range.

JC5607 will draw over 4A at 8.5V.

All of these bulbs are asking a lot from the AW17670 black label cells.

 

[Justin Case]

Yes, that's what I did, which I already said.

I must have misunderstood then when you stated

We don't know exact bulb specifications on SF bulbs, but we can deduce a lot from looking at current draw and comparing that to discharge graphs of CR123s and working backwards.

It sounded like you were using discharge graphs for 123A primaries, correlating those graphs to measured tail currents, and estimating Vbulb.

I'm talking about correlating the tail current to a bench supply V/I draw to estimate Vbulb.

 

[donn_]

Hi Donn,

Keep in mind, the 10.8V setting on the PhD-M6 provides soft start, but doesn't really provide meaningful regulation at the high loads presented by the 1185 and similar bulbs. Depending on the temperature and condition of the cells, the regulated run can be anywhere from non-existent, to 5-10 minutes worth.

You're ready to remove the bezel of the turbohead in order to install the 5761? They don't usually fit through the hole. Just making sure Also, at 7.2V, that's about 40W, which is pushing the cells even harder than the MN21 at ~35W, I think that's a bit too much for safety, unless it's not used much or only used on very short bursts.

It will be short burst use only, and most of my use for this M6 will be with Mag heads.

The 5607 at 9.6V: That's a 3-4C load on the cells which is very unsafe. I can't even guarantee that the PhD-M6 will be able to turn that bulb on. The protection in the cells will likely prevent it.

The 5607 is delegated to either level 1 (7.2V) or 2 (8.5V).

I think Will and I are going to have to discuss putting some kind of warning or disclaimer or something in place that says we are not going to configure the PhD-M6 in scenarios where the intention is to exceed 35W power consumption.

Eric

PS: when IMR17670s come out, I don't see any problem with powering up ~50W bulbs.

I expect to see IMR, LiFePO4 or some other safe high amp chemistry in this size eventually.

 

[donn_]

All of these bulbs are asking a lot from the AW17670 black label cells.

From AW's sales thread:

The new Protected 17670 (1600mAH), 17500 (1100mAH), 14500 (750mAH) are now available. A few improvements have been incorporated on these new cells :

- a new protection PCB with a higher current threshold ( up to 5A +/- 0.5A )

 

[mdocod]

It sounded like you were using discharge graphs for 123A primaries, correlating those graphs to measured tail currents, and estimating Vbulb.

I'm talking about correlating the tail current to a bench supply V/I draw to estimate Vbulb.

The results are the same either way if proper care is taken in adjusting for expected resistance and load. The drive levels for SF bulbs have been ball-parked on CPF since as far back as CPF goes. The PhD-M6 does not represent the first time that someone has tried to put numbers on paper about these bulbs. I've been figuring and estimating these bulb characteristics for years and building on that information and improving on it through those years. I've also conducted V-Bulb tests under the power of CR123s into SF bulbs on several occasions in the past, long before the PhD-M6 was ever even a dream.

Eric