Resistance is Futile (mag spring resistance)

click pic for the album of pictures.

I've been wondering a long time just how much resistance the mag spring has.. since even in my baddest (well most of them) mag mods i still use the stock spring.

So i finally took a miliohm measurement (the standard way.. jam an amp of current through it and measure the mV drop)..

the results... FORTYFIVE MILIOHMS! holy cow! That is more than double the entire resistance of my hotdriver circuit! Definitely more resistance than the smartpak and 'holy yikes' in general.

I took a piece of 1/8th inch wide solder wick and not even soldering just wedged it between the spring at the top and the bottom.. and it chopped off 2/3 of the resistance.. down to 15 miliohms! (0.015 ohm).

So.. guess i'll be performing that mod in the future on all my lights... I'll probably use the scrap wire i pull off the KIU ceramic sockets.. 16ga wire is 1/3 miliohm per inch.. So that will be perfect..

let me see.. with a 100W light and 9.5A.. that means.. 0.42V dropped on the spring.. that boils down to 317Lumen loss on a 64625 lamp!

anybody know a silver-plater?

I've been wondering for a long time 'cause steel is just a terrible conductor.. there seems to be some coating on them but apparently not silver (what it should be).. I'm surprised it took me so long to actually get the measurement. I've been just thinking that it's a pretty thick wire even if it's steel.. amazing once the numbers are actually calculated..

re-working for a mag85: 3.35A x 0.045ohm = 0.15V.. that's a difference of 38 lumen... 10.65 vs 10.80V.

Well one more thing to mod, folks... get out your soldering irons.

(oh.. ps.. sand first.. you'll not have much success soldering to that spring before sanding!)

-awr

I've been using solder wick on the springs for some time and it does help. Thanks for the numbers.

When I saw this I thought "no way is it that high", but looking at Wikipedia one foot of 18 guage copper wire is 6 milli-ohms, (and from a different page) steel has about 10 times the electical resistance of copper, so 60 milli-ohms per foot.

I'm guessing at the guage and length of the Maglite spring but this is in the ball park of what you measured.

Greg

and.. the spring goes round n round n round.. i didn't measure how long but a 'direct path' with some flexible wire is 1/3 to 1/4 the length for sure.. the 16ga scraps cut from making a KIU will be perfect for fixing the spring.. i'm just going to solder them right on the ends of the spring on the top.. on the bottom.. haven't decided if i'll solder on the very inside so it still has to go through a length of steel but is much easier to make.. or deal with cutting off some of the inner spring (what i'll most likely do) to make that easy to solder and shorter distance through the spring.

yes.. i was a bit amazed at first myself. I re-did the measurement several times.. i didn't try different springs.. some might have more conductive coatings than others.. but a real 'fix' is needed.

The solder-wick on the spring in the switch drops 80miliohm from the circuit. Glad to hear that somebody (bear) has been using this fix already.

-awr

Maybe this is why I haven't instaflashed my 9 x AA Mag85's bulb even when I pop in freshly charged NiMHs. I wonder if shorting the spring would remove the capability to use freshly charged cells. But it's tempting because of the rather large loss in lumens mentioned here.

andrewwynn said:

re-working for a mag85: 3.35A x 0.045ohm = 0.15V.. that's a difference of 38 lumen... 10.65 vs 10.80V.

Click to expand...

Have you actually measured this or are you going off of the LED operating curve?

That's why I used a foot in my calculation since the spring length is closer to a foot than to an inch.

Greg

andrewwynn said:

and.. the spring goes round n round n round.. i didn't measure how long but a 'direct path' with some flexible wire is 1/3 to 1/4 the length for sure.. .

-awr

Click to expand...

One thing I tried a few months back was first grinding the small end of the spring flat and then tinning both it and the bottom of the spring with some silver jeweler's solder - I don't know what the percentage of silver is because I bummed it off of a jeweler friend, but it's supposedly at least 10% - it melts so fast, and is certainly higher silver content than the common 2% silver solder carried at most electronics supply places. The plating on the Mag springs looks like it's zinc-based.

What about soldering the spring directly to the tailcap? Or, soldering copper caps to either side of the spring?

I always wondered if there is much resistance in the Mag tube where it makes the transition from the body to the tailcap - the parts are anodized, and I don't know if that adds a significant amount of resistance.

Anyhow, interesting thoughts in this thread. Thanks for bringing this up.
Chris

Well, after thinking about how my message appears, putting copper or silver on the existing spring will do nothing for the resistance of the steel. Duh! It did make for a less-intermittent connection for me, however.

You could connect the copper caps with the wick or other wire ensuring a copper path to the battery pack.

But I am curious about the tube-cap resistance, if any.

ciam... yes it gets a bit dicy with direct drive and instaflasshing .. like the http://prfix.rouse.com which drops the switch resistance to 1/6th a stock switch.. if you do that you'll be more likely to instalflash if you use an unregulated light. You do also get more bulb life if you leave the resistance in of course.

amanichen.. incandescent lamp, not LED, but yes using the formula.. but it's accurate... the formulas are extremely accurate for less than 10% voltage difference.

greg.. yes i was adding to your comment about the 'per foot' saying that you get close to a foot once you go round and round and round.

-awr

rook.. probably doesn't hurt to tin the ends, but it won't make a huge difference.. you are talking about changing 5mohm to 3mohm or something.. percentage big, total small...

The part where the tailcap touches the body is machined free of anodizing of course, else there would be no current flow at all.

Can't solder steel to aluminum, but it's a moot point anyhow.. there is so much surface area.. the contact resistance is minimal.. there is 10-20x the resistance in the spring as the spring-bat contact and maybe as much as 100x or more the resistance in the spring compared to the spring-tailcap contact (with that much surface area.. the contact resistance is probably 1/2 of a miliohm..

Also.. the body is an extremly conductive device.. 2.73 square centimeters of aluminum.. that's enough to carry 300A.. maybe more.. the resistance in a 2D body is.. 0.0000233 ohm.. no wonder people use alum. for flashlight bodies.

I always use pro-gold on the contacts such as the tube-cap connection.. i have actually 'burned' that contact before by opening a tailcap on a high-power light that was on. Fortunately.. surface area is the best thing to help a contact work better and that is a very large surface area... so will be extremely low resistance.

Oh.. yeah another thing.. i always sand down my springs 'til it's a flat surface else it just cuts into the battery and of course reduces the surface area greatly making it a much higher resistance contact.

-awr

Another thread concerning the same idea from a few days ago.
https://www.candlepowerforums.com/threads/129582

amazing timing on that post.. i just posted over there too.

-awr

My Fluke77III DDM doesn't measure sub-ohm resistance well, but it showed 0.1ohm for touching the probes together and 0.2ohm resistance from the top to the bottom of the spring. I'll see if I can borrow a better DDM to try to get a reading.

not sure if the 77 does 4-digit.. the 87s you can hold a button for a second to get 1 extra digit of precision. you can also zero the reading which is important. It is not possible to get an accurate reading from simply direct-reading.. but you can do to get close is this.. run a load like your flashlight through the spring (using jumpers etc).. and measure the mV drop on the spring.. now measure the current w/o the the spring.. the spring should have a similar resistance to the meter loading.. so you will have both a mV drop and a current.. now just divide to get resistance.. it won't be perfect or as accurate as my measurement where i put 1.00 A through the spring and measure the mV drop but it will be a good ballpark figure.

-awr

I tried a Fluke 79 but got the same result (0.2 Ohms in the probes, 0.3 Ohms from top to bottom of the spring in the tailcap). I'll try to find a Fluke 87 to use.

To measure currents and voltage drops I figured I'd need to get a good electrical connection with the tail cap removed from the light. Note that it is a cmacclel clicky tailswitch. The light is a cmacclel sub-2D ROP with a tailswitch: https://www.candlepowerforums.com/threads/105826

While playing around with the tailswitch, I tried to measure the total resistance of the switch in the ON position. I noticed that I couldn't seem to find a good point on the tailcap that would give me a low resistance reading from the end of the spring to the tailcap body (the threads, etc.). Where is the electrical path on a M@g tailcap supposed to be (through the threads or the top rim of the tailcap?) and what are the tricks to reducing the resistance along this path. I'm remembering reading about lye and such but I'm not a M@g mod or hotwire expert, that's why I got a turnkey ROP. Is there something I should be doing to the threads or tailcap rim?

I am wondering if I might be getting more resistance from the tail to body connection than from the spring.

https://www.candlepowerforums.com/threads/105826

andrewwynn said:

I've been wondering a long time just how much resistance the mag spring has.. since even in my baddest (well most of them) mag mods i still use the stock spring.

So i finally took a miliohm measurement (the standard way.. jam an amp of current through it and measure the mV drop)..

the results... FORTYFIVE MILIOHMS! holy cow! That is more than double the entire resistance of my hotdriver circuit! Definitely more resistance than the smartpak and 'holy yikes' in general.

I took a piece of 1/8th inch wide solder wick and not even soldering just wedged it between the spring at the top and the bottom.. and it chopped off 2/3 of the resistance.. down to 15 miliohms! (0.015 ohm).
-awr

Click to expand...

Thanks for sharing your measurements with us.

The MagLite spring resistance (and in other lights), has been a well known factor for a long time, as well as the solder wick trick. Another technique is the copper endcap and flexible wire found on motor brushes, just cut off the graphite rod. It can look a bit more professional, and they are low cost from certain sources (hardware store will gladly empty your wallet on these). It is good to remind the new folks around here about it...oh, and in certain other areas where one finds springs, you can often find a pretty decent drop, especially for those fine springs found in switches, where someone used it to carry the current...

NewBie said:

Another technique is the copper endcap and flexible wire found on motor brushes, just cut off the graphite rod

Click to expand...

I am trying, but I can't visualize this. Do you have a photo so that I can "see" what you mean by using these as alternatives?

Thanks in advance,
Will

wquiles said:

I am trying, but I can't visualize this. Do you have a photo so that I can "see" what you mean by using these as alternatives?

Thanks in advance,
Will

Click to expand...

Here is one that is for a small motor, for an example. This one here has a brass cap on the end:

Newbie's talking about the braided copper wire used for the likes of brushes in motors.. it looks just like solderwick but without the flux so it will probably not soak up solder so bad when you try to solder just the ends... a main issue with using solder wick for this type of mod is that it will become stiff with solder quite a way from the end as you solder.

The professional look i will have on mine is not using bare wire but a nice clean 16 ga flexible wire that will be soldered right to the end of both sides of the spring.. in-line with the ends of the spring and making a nice short coil down through the spring.

The spring inside the top of a magswitch that allows the goofy moving of the lamp by the patented quick-focus (useless IMHO).. mechanism is at LEAST 75 miliohms.. thin steel spring.. like 24ga.. it's horrendous.. check out http://prfix.rouse.com to see the fix that drops the resistance to 1/6th.

In a VERY related post.. check out my calculations here

-awr