Does the eGo have a single point failure built in?

[Rocketman]

This schematic has been posted elsewhere in the forum and after looking it over there seems to be a possibility for a single failure causing an eGo to go thermal.
I can't get into the Kitchen/Lab/shop at the moment because the wife is cooking dinner or I would test my assumption. Maybe someone can confirm this:

The Mosfet, Q1 in the circuit below is controlled by the eGo 'Magic Chip'. The gate turns it on and off. The eGo charger, when working properly will stop charging at 4.2 volts. If the Mosfet is OFF, and 4.2 volts is applied by the charger nothing happens. If 4.7 volts, or above is applied, doesn't the substrate diode iin Q1 start to conduct and continue to charge the cell?

If the voltage output from the charger goes above this point due to it's output Mosfet shorting (common failure mode for Mosfets) will the output go to 5 volts?

This looks like (at least to me) to be a single point catastrophic failure point for the eGo and Charger. Of course the eGo Mosfet, Q1, fails quite often usually giving a no-vape, or continuous vape indication, but if charging voltage is limited to 4.2v it will only trickle charge at 4.2 volts if left on the charger (not good) but even a good eGo will conduct if the charger fails voltage regulation.
How about some comments? Will the cell charge past 4.2 volts with over 4.7 volts applied?
Thanks.

EDIT: I get 0.5 volts drop across the eGo Mosfet in an OFF condition (positive applied to center pin, measured between center pin and cell positive).

 

[Tom10]

Would you be talking about a genuine Joytech, something that looks similar to a Joytech, or something that's obviously not a Joytech. I can only imagine this would be an extensive list and anyone's guess about what type of protective circuitry they have or don't have. Some of the clones/knockoffs look pretty good, and some of them actually work pretty well. Then there's the ones that are complete junk. I don't know how you would ever sort out all the data. Add all the types and generations of all of them (i.e. Ego-T, Ego-C, Ego VV, year made, etc.).

 

[Rocketman]

You think they use a different circuit design?
As far as I know even the junk ones, and the Genuiney ones use a PWM circuit with a single Mosfet controlling the output (and by default the charge input). Might be no sorting required. The substrate diode in a Mosfet switch is a fabrication feature). Even the Chinese protection boards used on Cheapo Chinese Li-ion cells have back to back Mosfets to prevent reversed diode conduction. That would require both a mosfet failure and a charger failure, 2 failures instead of one.

Did the knock-offs just copy an inadequately protected design?

How about only evaluating the PWM eGos and clones, not the unregulated RIVA and clones, not the VV eGos, eGo passthrus, just the single range, fixed voltage PWM models. The most common, and most copied of all the variations.

Tom10, what's your opinion on the Genuine Joyetech eGo circuit design?

 

[Rocketman]

I presently have an eGo on a 5 volt charger and it is getting quite warm.
About 5 minutes as of this post.

It's outside, away from the house, and we will see how long it will take to vent (if it does).

It is a known good eGo that I have tested the PWM output with a TeK TDS310 o'scope so I know the Mosfet is in good shape.

I thought about a video, but this could take a while. I'll post back if it vents.

at 10 minutes @ 5 volts it is too hot to hold (probably about 160F?). back on power again.
Past 30min and still getting hot, but no explosion, yet.
50 minutes, and the temperature has actually gone down some.
The button shows signs of starting to melt, but still no explosion.

Over an hour and the batt is still warm and drawing a little current from the charger.
This one is toast, but it probably won't blow.

The fact that it got too hot to hold on 5 volts in tells me overvoltage protection is inadequate or non-existent. A charger that fails and puts out over 4.7 volts (4.2 volts for the cell and 0.5 volts reverse bias on the mosfet) has a good chance of destroying an eGo.

It looks like something in this eGo has now failed and it is cooling off.
Darn

Hard to tell but it did get hot enough to start melting the button (seen next to an un-Rocketized eGo)

 

[Rocketman]

I disassembled the test eGo and all plastic parts were soft and mushy. The cell had about 2.5 volts, but would not load.
To my surprise, the circuit board still works (with another cell), but I'm sure it would not be trustworthy. I wonder if it stopped because of the excessive temperature.
Sure wish it had blown
http://i965.photobucket.com/albums/ae132/rocketman_photos/HotEgoApart.jpg

 

[Rocketman]

EGo heating up and goes into thermal runaway, some thoughts:

The eGo has an internal Li-ion cell, a small circuit board, some metal parts, and a charger usually connected to a low power source (USB).

If charge current and voltage is limited by the source,
there should not enough power input to the cell to warm it up. A normal eGo charger puts out less than 2 watts. That should not be enough to heat up the eGo to a critical temperature of 90C.
If the cell goes thermal, my guess it would be caused by charge over voltage, and then the cell chemistry.

Even a circuit failure inside the eGo would not cause an overcharge condition unless combined with an overcharge input. A Mosfet failure inside the eGo will not make the voltage go above the charge input. A short of the ground wire to metal shell would do nothing. The cell is wrapped and sealed. I don't think these overheating reports are due to internal shorts. Circuit paths from the positive cell terminal to ground are limited. The only low resistance, high current path to ground is the center pin in the connector.
In my test eGo, it went thermal (almost?) and nothing is shorted. Others that fail could be checked for shorts, or circuit board damage.

Obviously a cell failure could cause it's own runaway.

If a cell gets warm, something is wrong. A circuit board failure inside the eGo won't make charge current exceed what the charger can provide. Or make the voltage exceed what the charger is providing.

In my test, I was intentionally exceeding a safe charge input.

This cell I overcharged, did not end up with over charge. Did cell chemistry dissipate the charge and heat the cell? A permanent loss of capacity due to a chemical reaction as a result of high voltage and high temperature? Adding to the temperature rise? If the cell chemistry went exothermic, and was generating it's own heat, then the chemical reaction inside the cell could have exhausted the electrolyte. I guess I can try and charge/discharge the cell (outside again) to verify the cell has no capacity to charge and drive a load.

The circuit board now seems to work OK. The cell has almost no charge. I looked in the grass but found no charge left there. Where did it go?
I think I was close to success. Almost went BANG

 

[Janusxvii]

So if I am understanding what you are saying then exploding eGo batteries can ONLY be caused by either applying a potential voltage greater than 4.7V (i.e. using an improper charger) or a faulty battery? Naturally it is up to the end-user to make sure that they use the appropriate charger but does anyone know what Joyetech's Q.A. is like as far as weeding out bad batteries?

 

[Rocketman]

ONLY is not a good choice of words
I don't think many of the components in the eGo (plain old eGo) have much of a chance of making it go thermal.
The circuit is laid out well enough (I think) to prevent a short. Except at the connector. The positive terminal of the cell is at the top and is wired to a point on the corner of the circuit board. The low resistance path from there is through the Mosfet then to the positive wire to the center pin. Two surface mount resistors are attached to the positive circuit run.

The part with the highest failure rate is probably the series control Mosfet. But open or short, I don't see a charging issue with it. Open it won't charge, shorted it will charge until the charger stops. Unless the charger doesn't stop. A shorted Mosfet AND a charger that doesn't stop at 4.2 volts looks like a problem. That's two failures.

What I'm saying, or asking, is even with a good Mosfet. A charger failure that allows over 4.7 volts will continue to charge the eGo. 5 volts will take the cell to 4.5 volts. Not a good place to be.

Look over the circuit (if that is really an eGo circuit diagram). What single point failures can you identify that could cause a venting event?

The cell failing is an obvious issue. Does it cause most of the venting events?

 

[Rocketman]

It appears that the NEW version of the eGo-C has "improved over-voltage protection", per the JoyeTech website.

 

[Janusxvii]

Yeah it seems like it's just begging to be torture tested.

 

[Rocketman]

Wonder WHY the Upgrade eGo-C has improved over-charge protection?

 

[CountSmackula]

Very interesting, but it makes my head hurt trying to wrap my mind around what you're saying. Kinda like listening to the wife's cousin explain tube amps & other electronic gear. I should have studied more electronics.

p.s. Where in LA are you? I'm in New Iberia.

 

[Rocketman]

North shore of Lake Pontchartrain.

 

[CountSmackula]

How high did you have to raise your house after Katrina?

 

[Rocketman]

We didn't flood. Had more down trees than you could count.
MREs for about 6 weeks, Generator with limited gas, was sort of like camping.
Found out the kids really didn't like camping

Wore out a chain saw and bought a bigger one. I'm ready now.
Chainsaw, gas, eCig supplies, and ammo.

 

[opsguy]

It appears that the NEW version of the eGo-C has "improved over-voltage protection", per the JoyeTech website.

If I am not mistaken the two most recent reported battery failures where they exploded while on charger where eGo-C batteries. Makes me wonder if they truly have improved over-voltage protection or is it a QC issue with these newer batteries?

 

[Janusxvii]

If I am not mistaken the two most recent reported battery failures where they exploded while on charger where eGo-C batteries. Makes me wonder if they truly have improved over-voltage protection or is it a QC issue with these newer batteries?

I could be mistaken but the way I understood what I saw on the Joyetech website was that this new battery was a separate upgrade item not included in the kits. The kits would still have the standard battery.

 

[Rocketman]

I wonder why we don't hear about Rivas venting as much.
Market share?
They fail because of the Mosfet just like the eGos but any reports of venting?
I have 3 that were sent to me for parts, and they have Joye cells in them that look just like the eGo.

 

[JW50]

Rocketman - Interesting hypothesis and thoughts here. Wish I could add on the failure cause. But, a bit over my head. But have seen some other ECF members that seemed to have lots of knowledge in the circuits and mosfet areas. Wish they would enter some thoughts and opinions. Interesting test to reach "rocket" status. Maybe the conclusion should be if Rocketman can't make a rocket of an eGo then eGo must be safe from nearly any abuse?

 

[Rocketman]

Ha, yup if I can't blow one up - - - - -
I'm open to suggestion from the experts.
I don't own every model of the eGo, or clones that are out there.
The Joye site implies that the new eGo-C Upgrade battery has improved over-voltage protection, with a "device inside to prevent over-charging".

There is of course the other component that could create a single point catastrophic failure,and a few multiple components failures that could also. I'm used to system designs that don't allow single point critical failures.
Maybe I'm wrong on this one.


A little more thought. EGo, or clones, that vent violently while on charge, after being on charge for a while, not immediately, or those that get HOT to the touch, are probably being overcharged. Charge a Li-ion cell to 4.5 volt and it could go critical, generating it's own heat. Unlike mods that don't have undervoltage protection the eGo cuts off at a cell voltage before it get to 3.0 volts. Unless the control chip dies. Users usually vape until the eGo stops, and flashes at them. The white or blue LED reguires at least 2.8 volts to illuminate weakly, 3.0 to illuminate with normal brightness. So they probably aren't being over discharged. It could be faulty cells. But a faulty cell would likely vent under a vaping load.

I think the most likely failure mechanism for a Li-ion cell venting after being on charge for a while is over charging.

So, to the experts:
Would a faulty 5 volt charge input to a good eGo charge it over 4.2 volts; and if not, what component blocks that voltage from getting to the cell?