Improving the reliability of twistys

I think I created a failure mode, and a fix for tail twisty switches which might be useful for other twisty switches. I have these Xnova 8 "reversed" lights that have tail twisties. The primary mode of contact is that a plate in the tail - otherwise isolated from the light - shorts the battery to the back of the body when the tail is screwed down.

The back of the body has some burrs so I stoned it down to give a larger contact surface. Rather than improving things, this made the contact worse - weaker, with flickering.

Looking at it, I realized that the tolerances for this light were pretty tight: the bottom of the tail piece butts into the body just a little after the internal tail plate shorts the battery to the body. Because I had removed some metal from the back of the body, the tail piece thread now bottomed out at about the same time as the tail plate closed the circuit, making the contact iffy.

So, I ground a millimeter or two off the bottom of the tail piece itself (easy - the metal is thin there). It now remains well clear of the body so the internal tail plate makes a good, solid contact when it is screwed down. NOW it is reliable.

I'm thinking that the same fix might have been applied to the X1 I used to have and other twisty tail switch lights that got flickery.

If you don't mind the appearance, it might also be applied to head twisty lights - like the old CMG Infinities which have deformed their batteries so the switch bottoms out before the circuit is closed. It saves putting that aluminum foil ball into the tail.

Lights like the Dorcy 1AA and 1AAA lights don't have a body step limiting tail cap travel, so this is not a problem with them and mine have been 100% reliable with lots of use.

Just a thought. BTW, I've now improved my Dae type Xnova 8's with bored out LED holes and the ground down tail switch. Loaded with a BS lithium primary, using the stock electronics/LEDs, it makes a kick *** $4.50 shipped light, and I'm now in the process of freezing and popping the rest of mine.

I also had a contact problem with my Xnova8 once I "fiddled" with it. The battery tube is used to make electrical contact (positive), but I noticed that both ends were covered in the colored coating and thought that might be limiting the current (raising resistance) a bit. So, I sanded down both ends of the battery tube thinking it'd make it a bit brighter or something...ended up the light wouldn't turn on any more unless I tightened the tailcap on really tight. Like you said, it has really tight specs and it was actually counting somewhat on the additional length the colored coating offered to make the electrical contact. My solution was to add some solder to the copper traces on the tailcap PCB where it contacts the battery tube and it's now fixed with no contact issues.

I don't own any Xnovas but I would think twice about doing anything to a twisty that would shorten it, particularly if you use rechargeable cells. Some stock twisties can compress the cells enough that they will no longer be chargeable in some chargers or useable in some devices. Some models have been reported to noticeably shorten cells after just one discharge cycle.

Even though this is only an issue with some models of twisties, it's really a drag because the damage to the rechargeable cell(s) is only noticeable after it's already done.

I'd like to see a chart of twisties that always shorten cells.

Sub_Umbra said:

..Some stock twisties can compress the cells

Click to expand...

There are 2 types of twistys: the "vise" type which just push in 2 contacts points, and the spring loaded type which are always touching the battery, and the twisting action closes the circuit via a separate contact plate, usually in the tail.

I agree with you with the vise type - I was probably premature suggesting filing for a CMG infinity. I think filing is OK for the spring type - like the Xnova 8, Gerber LX's, etc. - as filing shouldn't affect pressure on the battery for these.

There's another factor that may cause damage to cells in either type of twisty. In an effort to be just a tiny bit shorter in overall length than the competition's light some lights have battery compartments that are just too short by design. These lights will shock and shorten cells that bottom out forcefully if the light is dropped in a mishap because there is just not enough room for the springs/contacts to absorb the force gracefully. Of course, any light will deform it's cells if dropped from high enough. I'm referring to reports I've read here of damage from short drops. (2-3 feet)

Unfortunately, the designers that suffer from makeitalittlesmalleritis also make the same error with some clickies.