Has anyone measured and/or considered the electrical resistance of the battery's spring contact?
With high current drivers it's crucial to maintain very low the electrical resistance of the power path. As an example, if you have a good battery with 0.1Ohm of internal resistance AND a bad spring contact of 0.1Ohm we have a voltage drop of 0.6V at 3A and a power dissipation (lost power) at the battery contact of 0.9W...
Has anyone thought at this problem and found an optimal solution?

Clean the battery contacts and the springs. And use gold-plated springs for the build. You can't get it any better.

to reduce the contact resistance, the contact surface should be maximized...
any proposal?

I doubt your battery spring has much more resistance than the traces in your driver's PCB, probably much less.

Those springs are like 24-28 gauge? At 28 gauge, you're looking at about 0.065 ohms per foot according to this table, so the spring probably has about 0.01ohms of resistance, if that.

...no, the problem is in the contact between battery's electrode and the spring, not the spring's electrical resistivity...
moreover, you can't use the table because it's for solid copper conductors and the spring isn't made of pure solid copper.

i would call the losses, negligible.

If you try to measure the voltage at driver's input when you are draining more than 3A from battery you could have a bad surprise...
...multiple voltage drops at battery's contact, battery's switch and aluminum hosts ground contacts...

Every spring I've seen has greater contact with the battery than the wire's thickness. I would consider these losses negligible. It's not like we're using nichrome or something. Feel free to experiment yourself and report back on what you find.

laserluke;

Contact resistance becomes important as amp draw increases in our lasers.

Gold-plated contact springs soldered to circuit boards is a good practice:



I find switch design to be another source of high IR.

LarryDFW

P.S. Have you made any metal-core PCB's for resale yet.

Not yet, i'm waiting for a general interest about this driver to collect more requests and thus reduce the overall costs and time.
Another way could be an agreement with someone that does the job to manage more requests...

Why gold-plated? It doesn't improve conductivity, and nickel works fine for corrosion resistance. These losses you're worried about are negligible. I bet it accounts for less than 2% of total power losses in the system.

I have no idea what you're trying to say here.

He means Internal Resistance, not Infrared.

I've checked a few tailcap springs/clickies and the resistances have varied between 30mOhms and 330mOhms. That's not negligible. All of the good ones in my builds are <150mOhms, and I'm betting most of that is due to internal switch design and not the spring itself. The poor ones had poorly plated (not gold or nickel) springs or unplated/oxidized springs. Some of the tailclicky switches rely on two contacts touching less than 0.25mm^2 of surface area with very little pressure, this too could be a major source of resistance if oxidation or dirt is present.

I think the main problem is that no matter what you do, you can't guarantee that it will increase the actual contact surface area. One of my lights has a brass puck mounted to the top of the spring to make the actual contact, and I've soldered the spring to the puck to ensure good contact there. I think one good way would be to solder contacts to the batteries and use low resistance connectors, but then you lose the convenience of an easily changeable battery.

I've seen some people solder some desoldering braid to the spring in such a way that it bypasses the spring, but as others have said, I don't think the spring is causing the resistance, I think it is the bottleneck of the small surface area of contact. I found a spring where the ends are flattened, i.e. it's not round wire all the way around the spring, rather the wire ends up flattening out at the part where the coils get closer together, so that when you stand it up on a table the entire bottom of the spring is touching the table. A spring design like that ought to be helpful, maybe.

Oh, and one more thing. Since it is so hard to solder to aluminum, we almost always rely on pressure for connections, whether it be from a retaining ring or from bolting a wire to the aluminum. IIRC, RHD found some flux that will allow you to solder to aluminum. Using this and solder connections would more or less rule out those junctions as the source of resistance.

Oh, and one more thing again. You can buy lower resistance clickies (I think the mcClicky is lower resistance, but don't quote me on that), and even electronically switched clickies such as the FETtie. Assuming you make the connections solid, the FETtie should offer the least resistance.

In many builds we used a blue diode with 1.8A and a forward voltage of about 4.5V and so 8.1W. With a driver's efficiency of we say 75%, we have 10.8W at supply input (2.91A at 3.7V nominal battery voltage).
If we are too much lucky and the total spring-aluminum-switch resistance is only 0.1Ohm, we have a voltage drop of about 0.3V. So the driver sees an input voltage of 3.7-0.3=3.4V and increases the input current to 3.2A to compensate. And so the final voltage drop is even greater...
Should we still think that it's negligible?

who has measured their spring resistance? and/or voltage at the driver Vs voltage at the battery?

I just measured a UV light I built with, and without the switch assembly.

It draws ~ 1 amp of current.

The switch I tested is from a well-designed host
(the one in signature below).

It has a large switch with a soldered large spring,
and a threaded ring contacting a PCB which is soldered to the switch.

The switch as received dropped the operating current to ~85% of the current with a jumper across the battery cell negative post to the host battery tube.

Further testing with some conductive wire compound showed that the contact resistance could be improved to about 90% of the direct jumper operating current.

Bottom line is switch & contact resistance in the base of most hosts . . .

IS LIMITING the power delivered to our lasers and lights.

LarryDFW

Resistance in mechanical contacts like battery springs, but also switches and plugs, is always a tricky thing. One important factor is contact pressure, i.e. it helps if there is a lot of force pressing down on the battery.

On higher currents (over, say, 1 amp) its not unusual for the contact resistance to be quite high when you just insert the battery, but it will get much lower once you turn the device on. The contact resistance causes local heating which can sometimes more or less solder itself in place. This usually isnt visible or noticable when you remove the batteries, but it does happen on a microscopic scale.

You can try measuring the resistance before and after turning the laser on after you inserted a new battery.

Use a constant-current driver next time

Constant-current drivers would help in some situations . . .

but if too much voltage is lost,

even a 7135 or buck driver type will drop current.

LarryDFW