Why LED torch flickers with RCR123?

[cykoed]

Hi,

I bought the following LED torch and some rechargeable CR123 batteries
[http://www.dealextreme.com/details.dx/sku.13338]. Silly me...I didnt read correctly that the torch only specified CR123A batteries and did not include rechargeables

- Model: Smart-Fire A-1
- Features a Cree XR-E P4 LED emitter
- Powered by 2 x 3.0V/3.6V CR123A batteries
- 150 lumens max (manufacturer rated)
- Single-mode clicky tailcap switch
- Glass lens with textured/OP aluminum reflector
​

The torch flickers continously on the RCRs but works fine with primaries.
Question1: Why does it flicker with RCRs?
Question1: What can I do to prevent it flickering on RCRs?


Rechargeable batteries that I bought

• Ultrafire 3.6V 880mAh LC 16340 Protected CR123A Battery 2-Pack [http://www.dealextreme.com/details.dx/sku.3273]
• TrustFire Protected 3.6V 880mAh 16340/CR123A Batteries (2-Pack) [http://www.dealextreme.com/details.dx/sku.8683]

 

[WadeF]

Well, when you purchase cheap lights from DX (I have bought a few myself) you run the risk of having problems like this. Not sure why it is flickering, probably something wrong with the circuit. You could try sending it back to DX for a replacement.

Also make sure all the contact points in the light are clean.

 

[cykoed]

Wonder why it flickers with RCR but not with primaries?

Well, I'm waiting for a replacement bulb for by SF E2E on which to test the RCRs in case the batteries are the problem.

 

[koala]

Wonder why it flickers with RCR but not with primaries?

Well, I'm waiting for a replacement bulb for by SF E2E on which to test the RCRs in case the batteries are the problem.

Don't try original SF E2E bulb on RCR123 they will blow.

 

[cykoed]

koala, unfortunatley i found out the hard way and blew the SF bulb

getting a replacement bulb from lumensfactory...waiting for it to arrive. they advised me that their bulbs are ok to run on rcrs.

 

[richdsu]

IMO the circuit board is faulty.
When you use RCR battery, it has higher voltage and it will generate more heat than CR123A ( depending on the circuit - buck probable) during regulation.

If the components on th circuit board is not properly soldered, and if there is hairline solder cracks, when heated up the circuit will perform erratically.

Another cause may be faulty component not able to perform properly when it is heated up.

 

[AvidHiker]

IIRC, its not unusual for some drivers to flicker when fed voltage exceeding their normal operating range (like the ~4.2v from a fully charged 3.6v RCR-123A). In some cases, the flickering will stop once the cell has discharged a bit.

 

[Justin Case]

The higher voltage of the 3.7V (typically 4.2V fresh off the charger) RCR123A cells most likely exceeds the voltage range of the LED driver board. When that happens, flickering or strobing is a common result.

You can stick with Li primaries, go with 3.0V rechargeables (but have to accept drastically lower mAh capacity vs. 3.7V cells), find a different drop-in lamp that can be driven by two RCR123As, find a different driver board that works with two RCR123As, or get rid of the flashlight and find another one.

 

[MrGman]

+1 for Justin Case's explanation. Its just outside of its (the driver circuit) voltage regulation range with the RCR batteries. I have the same problem with the Solarforce R2 multi mode pill on the lower modes. If I use good freshly charged RCR's the light doesn't flicker in high mode because the current draw pulls the voltage down just enough, but in medium mode and low mode it flickers and I can hear it. At the lower current draws the voltge goes back up and its just at the margin of the regulation range so it flickers and makes audible humming modulation sounds. Goes away every time with lower voltage batteries.

 

[bessiebenny]

The specs clearly say it supports both 3v and 3.6v CR's. So that includes RCRs.
I highly doubt it has anything to do with voltage.

It should most likely be due to the length difference of the batteries.
Most protected RCRs are quite a bit longer than primaries in length.
So the contacts may be affected. (Spring gets too condensed ot threads don't close as tight)

eg. Many lights including various FENIX lights have issues with protected RCRs.

btw. Ultrafire RCRs should be shorter in length than Trustfires.
Try the Solarforce (ebay) or KD Protected blue RCRs. (kaidomain)
These are shorter and have been showing very nice performances.

 

[Justin Case]

3.6V is the max open circuit voltage for Li primary cells. Under load, however, the voltage is typically less, say 3V and under. With Li secondaries, the voltage under load is around 3.6V-3.7V. Hence, the voltage range of the driver is probably exceeded when using Li secondaries, resulting in flicker.

 

[bessiebenny]

Yes. Of course I know that they can be as high as 3.6v.
(Although Panasonic's are 3.0v exactly and most are only ~3.3v)

But. When it says 3.0v and 3.6v CR's in the specs, to me, it 100% supports RCRs. (RCR's are 3.6v but charged up to 4.2v)
By the way, it clearly says it supports rechargeables on the back of the packaging so no need to guess. =)

Also. Many other lights say 3.0v to 3.6v CRs when it supports RCRs such as:
http://www.dealextreme.com/details.dx/sku.14068

- Powered by 1 x 3.0V~3.6V CR123A battery (sold separately)
- Voltage input range: 3.0V~4.2V

 

[Justin Case]

You seem to be placing an inordinate amount of faith in specs on a Chinese web site of dubious command of the English language. Even a broken clock is correct twice a day. The DX light you cite could be one of those specs that are actually correct. As for stated support for RCRs, those could very well be the 3.0V RCRs, not the 3.7V ones.

 

[bessiebenny]

I guess it's due to having many experiences with these cheap lights and know what to expect. =)
(many, I mean many many of these lights all use very similar drivers from same factories)

 

[MrGman]

I still have to agree with Justin Case, I don't see how the fit of the RCR batteries would have anything to do with the flicker only in certain modes of operation. BB you are putting way to much faith in the text of an advertising claim. Not all well made analog electronic circuits (or digital for that matter) behave the same with various components, you expect a low budget regulator circuit made in China to have optimum performance because it says so on a piece of paper or website?

The voltage to the regulator does in fact change as the current draw goes up and down. At lower load currents the voltage goes up and may be right on the margin of what puts this thing out of regulation so the current vascillates back and forth. Putting it under a high load current which causes the voltage to drop brings it back into regulation and the problem goes away. This is a common problem with any power supply design where the input voltage source is allowed to vary too much. Most actual regulator chips have a simple voltage comparator inside that does all the work of sensing 2 lines for differential voltage drop and changing the output to drive something accordingly. And even those chips have tolerances on bias voltage, offset voltage, drift with temperature, so on and so forth.

I have seen it in for years and years in various power supplies and I have seen it in my Solarforce R2 5 mode pill, stictly as a function of bringing the total voltage between 7.6 to 8.4 V.

I can run it off of 2 low budget RCR123's that don't have much umph behind them (current carrying capacity) without flicker, But I can't run the exact same module off of 2 freshly charged AW17500 cells using the cell extension with the Solarforce L2 host. Has nothing to do with fit, they all fit great. In the 2 lower power modes it flickers very badly with these better batteries that hold up the voltage better under the lighter loads.

Not everything advertised is exactly what it says it is. All components have tolerances and when you put the tolerances of a series of components together in one circuit you have tolerance stack up and no 2 devices have to behave exactly the same. If the design tolerances were marginal, then so will be the performance when in this case the voltage range is pushed to those margins. This is nothing new or even exciting. If he went through 100 modules he would probably found some that would accomodate his batteries better and some that don't, but a consumer buying a single copy of a product doesn't normally get to see that.

Justin Case, do you have a cousin named Justin Time? :thumbsup: Sometimes uses his middle name thenickof?

 

[bessiebenny]

I don't expect optimum performance. It's just how it is in most cases so far.
Pretty much every "chinese" lights support RCRs coz that's the "cheaper" way.
(obviously there are exceptions still and this light could be one of them)

Also, I know that there are 3v rechargeables out there. I have them. LifePO4 ones.
But they are not common. So I doubt they wrote "rechargeables" on the box while thinking about them.

But yes, there is a high possibility that they just wrote down the specs wrong. =)
I guess we won't know the answer until someone confirms it at DX or if I buy one myself and test. hehe.

- I've bought/used/reviewed over 50 different LED lights in the past 9 months or so and most flickered when it had bad contacts.
- Only had 1 flickered due to a faulty driver. Had none so far which flickered due to using RCRs however.

So I'm just giving my oppinion to the OP based on my experience.
It may not be always correct but more ideas, the better it is imo.

 

[AndyTiedye]

You might consider LiFePO4's if you want to run rechargeables and the light doesn't like regular li ions.

LiFePO4's have similar voltage characteristics to lithium primaries.
They also will not even if abused.

 

[Justin Case]

The very first Chinese drop-in I bought -- Kaidomain SKU 2363 -- strobes/flickers when driven by two 3.7V AW RCR123A cells (and the KD web site clearly states that the 2363 is not rated for two RCR123A cells). Does not flicker with two standard SureFire Li primaries.

In fact, the KD2363 also has sort of the "reverse" problem in addition to the above strobing behavior. When driven by one SureFire primary, the output is poor. When driven by one AW cell, output is normal.

So, your experience is certainly not universal nor a law of physics.