Roomba 880 melting charging contacts

[Gordon]

Some more detail pics of my above mentioned Roomba+home base. ...

I am not familiar with the scene in your lower photo, is it fair to say we are looking at the mains-receptacle somewhere on your Home Base? If correct, where might that small, loose, and apparently damaged piece have originally been located?

[Gordon]

... Roomba and docking station plus contacts melted. ...

Damaged from HEAT, yes! But I take exception to the way the description "melted" is used in this thread (not your doing--of course). Most of what is seen in this thread is movement of softened plastic, having been pushed out of position by applied mechanical force. OTOH, your Home Base enclosure had such severe local heating that the plastic slumped due to the pull of gravity--detailed in this vignette of your photo:

The slumped zones appear to me to have formed by this HB-contact getting hotter than its robot-mate, to then heat, without contact, the adjacent HB plastic. The plastic was at a temperature verging on actually "melting"--going into the liquid phase.
To introduce a general response of heated thermoplastics, the progression goes something like this:

• 1) A thermoplastic casting undergoes controlled heating in a thermal-chamber.
  2) As the object's temperature increases, its strength begins to decrease.
  3) At some elevated temperature level, the object can be deformed via application of a force level that would make no permanent change to it while much cooler.
  4) Upon increasing the chamber temperature much further, and allowing time for the item to become isothermal, an observer may notice that only the pull of gravity is sufficient to distort thin sections. This status relates to the slump-test temperature of a thermoplastic.
  5) Additional increases in chamber & object temperature result in converting the plastic to the liquid phase. Its melting point has been reached.

Step #5 is what I think of as "melted". Much of the action reported in this thread equates to step #3 temperature. BTW, none of the talking points are exactly a single temperature number for a given plastic formulation, instead they are temperature ranges.

... For me it is a problem by design.

I agree with that statement, but am currently thinking it needs to be split into two topics:

• A) Structural & thermal design; AND
  B) Design specification of the plastic formulation used to make chassis and housing castings.

Making use of vic7767 and Brett's pix that show inner views of the robot's chassis structure where the charging contacts mount, I back tracked to a 2010 R3 (560) to confirm that (probably) iRobot had made no engineering change to those features (they were not even bothered when the chassis was modified to use a separate battery hatch!). It is apparent that no thermal consideration impacted that original structure; i.e., no heat transfer / sinking was needed. I tend to think the structure (HB-charging-contacts region) is constant all the way back to 2007, when R3-Roomba hit the streets.

I do not think this forum experienced any over-heating issues during the 5XX-6XX 'era'. If that is true, then we need to look at (B) for the current culprit.

If resistance gets higher and higher voltage drop becomes bigger and bigger so current raises to a value to melt the plastic parts. If they would use metal it would be no issue I think.

Maybe, all that is necessary is to use the same plastic formulation for robot chassis, and HB housing castings as was used from 2007 to ~2010, except dark gray or black in color.

Do you think this problem could end up in a more dangerous problem like a fire caused by the robot?

No, but with caveats. I think the excessive heating will become self-limiting. I'm looking for contact resistance to be increasing, and know that the voltage across / between robot-contact and HB-contact will decrease as battery voltage rises (with charge acceptance). Contact-resistance should increase due to surface spark-errosion, so the Joule heating power: deltaE^2/Rcontact should reduce while high-rate charging progresses. Any softening of the plastic supporting the robot's contact plate will permit the plate to move farther from the HB's base, thus reducing the applied force (from the mating HB's spring-contact) and further increasing contact resistance.

I expect the 7XX & 8XX Roombas still invoke a charging halt, with error, when current falls below 400mA for 30 minutes--but, of course, the heating game is by then over.

Another mitigating factor is: There appears to be no "molten plastic", expected to exude flammable volatiles, in reported events, hence that can add to 'comfort'. All said and done, I don't think I would sleep well with that bugger charging in my house overnight!

Regards

[harby]

Thermoplastics don't "melt" at a specific temperature, they just keep getting softer and then more viscous. I'm thinking of specific low temperature examples such as the thermoplastics used to mold connectors at the end of power and network cables, and the kind used by Hoover to mold fittings around the end of vacuum hoses - those can be made pliable and reformable at under 100C, as in "immerse in boiling water and reshape". The material goes from elastic to plastic at temperatures lower than 100C, and the deformation would become permanent.

That's a specific low-temp plastic though; standard injection molded plastics should have significantly higher temperatures before they yield inelastically. I suppose if one wanted to do empirical measurement and had some Roomba plastic to sacrifice, you could put a flexural test load on a piece and slowly heat it to determine the temperature these materials must have reached.

The power supply puts out 22.5V, that's always going to be higher than a 17V battery. Connect just 0.5V across a 0.4 ohm 1/4W resistor (for coincidentally 1.25A) if you want to see how hot small voltage drops can make things (wear safety goggles).

Additional shower thought: you mentioned arcing, which could produce it's own mini carbon resistor on the surface. As the terminals only have 5V until charging is demanded, the only case where this seems likely to happen in design is if someone lifts the robot off the dock while it is charging. A coating of dirt or debris, like the robot running over a cat-peed carpet, could also give the contacts a conductive but resistive film.

[Brett]

Classic you guys! Reminds me of one of my favourite Engineer jokes

A man in a hot air balloon realised he was lost. He reduced altitude and spotted a woman below. He descended a bit more and shouted, "Excuse me, can you help me? I promised a friend I would meet him an hour ago, but I don't know where I am."
The woman below replied, "You're in a hot air balloon hovering approximately 30 feet above the ground. You're between 40 and 41 degrees north latitude and between 59 and 60 degrees west longitude."
"You must be an engineer," said the balloonist.
"I am," replied the woman, "How did you know?"
"Well," answered the balloonist, "everything you told me is technically correct, but I've no idea what to make of your information, and the fact is I'm still lost. Frankly, you've not been much help at all. If anything, you've delayed my trip."
The woman below responded, "You must be in Management."
"I am," replied the balloonist, "but how did you know?"
"Well," said the woman, "you don't know where you are or where you're going. You have risen to where you are due to a large quantity of hot air. You made a promise, which you've no idea how to keep, and you expect people beneath you to solve your problems. The fact is you are in exactly the same position you were in before we met, but now, somehow, it's my fault."

As an engineer, I'm allowed to make engineer jokes!

I think it's fair to say that if you tapped a group of random forum members on the shoulder and said "what happened here?" most of them would say "It's melted!" and if you said "Heck no, it simply heated to a point of reduced yield and suffered non-elastic deformation under the force of gravity!" they'd look at you funny...

But, of course, you guys are right.

But focusing on the problem, the only thing we know is that there is undue localised heating at some charge tabs causing problems.

Things still to be proven (or disproven):
1. It seems to be only in the 220V/240V areas.
2. It seems to affect the left hand tab.
3. Seems to be prevalent in later models ie 7xx on.

Notes on 1: I have had this from Australian supplied (ie factory) 240V power supplies and the one I have in now is a fellow that was running a 120V supply on a local (plug-in) step-down device. Australia is nominally 230V / 50Hz.
Notes on 2: This is simply my observation from the 3 that I've had in and the pictures in this thread - small sample, I know, but it's there.
Notes on 3: One of the cases coming through here was a 6xx (see pics I added near the beginning of this thread - minor case though), but certainly two were 7xx and the one I have in now is an 880.

(-8

[GermanRoombaUser]

OK, forget about the term "melted". You are right, material became elastic and by force (at docking) we got plastic deformation.

I keep my statement, if they would use metal, this would be no issue. Of course, the general point is they better use materials (cupper, aluminum or "stronger" plastic) wich can only be deformed at higher temperatures. Maybe here it was only a bad badge of plastic parts or a general material change (e.g. "reduce overall cost -> reduce plastic production cost -> reduce energy consumption -> reduce melting temperature of plastics -> take a new material mixture")

The brown spots on my home bases contact I also think are made by arcs. But I guess this is a result of the original problem, not the cause.

[Brett]

No - I'm saying there shouldn't be any arcing.
There shouldn't be any local heating.
The current design should be fine (structurally anyway).

Something else is happening?

Dunno.

[bkmohs]

I posted earlier about two 880s with this problem. Both were returned to iRobot this week - they sent replacements under warranty. This is apparently a well-known problem inside the company.

Regarding the ongoing assumptions, it was the left contact that had deformed/retracted into the base on both, but I'm in the U.S. and on 120v power, so this problem is not limited to 220/240v countries.

I have no idea if the replacements they sent have been improved in any way (plastic reformulation or whatever), but consider it unlikely, so will be on the lookout for recurrence of the problem.

By the way, I did hear the tick-tick-tick sound on these units while charging off-and-on for several weeks, so am assuming that was arcing from the charging base contacts. And that this helped drive the overheating of the contacts and the softening of the surrounding plastic. My assumption is that the spring-loaded contactors in the charging base are what causes the charging pad on the bottom of the vacuum to recess into the bot.

When the arcing happens and the metal pads get hot, the surrounding plastic softens and the spring on the charging base pushes it up into the body of the bot. Now, WHY the arcing happens and overheating occurs is up for somebody else to figure out. I hope they do that soon, since this will be ultra-annoying on a regular basis.

If iRobot doesn't address this and it's a mass occurrence, then I smell mass recall and lawsuits coming.....

[vic7767]

I an putting out this question to rr members. It seems that the 880 is the first Roomba model with the PSU built into the charging base and it is universal i.e. both for 230-240 vac 50 cycle as well as 110-120 vac 60 cycle. There is no stand alone home base that requires an additional PSU be plugged into it. Could it be something in the new integrated PSU/Home base that needs a redesign instead of the 880 Roomba itself ????

[Brett]

... Could it be something in the new integrated PSU/Home base that needs a redesign instead of the 880 Roomba itself ????

This certainly appears to be a likely item on the iRobot "to do list", but I have had two 7xx and one 6xx with the same issues. So there must be enough in common in the circuitry to have transferred the problem from the old dock design.

If iRobot do a redesign on the integrated unit, I wonder if they'll make it backwardly compatible and supply it with the 7xx and other units that are still being sold?

[Conker]

Vic7767, I own a 2014 model 880 and there is no evidence of any problems with either the charging contacts on the Roomba or the base station. The front caster squeaks and the battery has lost some of its punch, but it still gets the job done.

As a boot note, I live in Japan where the interior rooms are often partitioned by sliding doors. The runner tracks on the floor do seem to take their toll on the rubber rollers, which will probably be the first things to warrant replacement. That and the tyres on the wheels.

Interestingly, the Roomba seems to detect my rabbit and slows down for him when he is in its path!

[gabrielleadams]

Exact same thing happened to my 880 this week. Melting around one of the contacts, which has sunken into the machine, plus melting around the contact on the charger. Thankfully, I caught it before it started a fire.

They (iRobot) are saying, "This is not something we see from the Roombas and we would like to have the Roomba come back for our people to look over."

[Brett]

Things still to be proven (or disproven):
1. It seems to be only in the 220V/240V areas.
2. It seems to affect the left hand tab.
3. Seems to be prevalent in later models ie 7xx on.

1 - ??
2 - CHECK
3 - CHECK

"1" should now read "It seems to be prevalent in 220V/ 240V areas." as there has been a US user with the problem in the mean time.

(-8

[robocleaner]

Classic you guys! Reminds me of one of my favourite Engineer jokes... As an engineer, I'm allowed to make engineer jokes!

I'm now retired. But I wish I'd heard that joke 20 years ago when I was in Management and used to train others... as with the best jokes, it's an observation on life, and it's too true to be funny!

I think it's fair to say that if you tapped a group of random forum members on the shoulder and said "what happened here?" most of them would say "It's melted!" and if you said "Heck no, it simply heated to a point of reduced yield and suffered non-elastic deformation under the force of gravity!" they'd look at you funny...

If it looks like a Duck, walks like a Duck, and quacks like a Duck... it probably is a Duck. A melted plastic Duck of course, but a Duck none-the-less.

[pa4wdh]

I own a roomba 880 for a bit more than a year now and i just checked if i could see indicators of this issue on my robot.
The contact on the homebase looks a bit like the picture Gordon posted, although the plastic is still undamaged in my case. On the roomba side i get the impression that the affected contact is a bit deeper inside the robot than the other, again, the plastic seems to be in tact. On the contacts themselves (both of them) i can see little black dots, but i don't know if that's related.
And for statistics: I'm using it on 230V/50Hz.

One thing i also noticed is that when the roomba is ready and docks back onto the homebase it doesn't always play it's "mission successful" tune. Could that be related ? However, it does charge when that happens, so maybe the contact with the homebase is not ideal in those cases ?

[GermanRoombaUser]

Can you hear a "tick tick tick" noise while charging?

[pa4wdh]

The homebase is located underneath my couch so i don't pay much attention to it, but i have occasionally heard a ticking noise while the roomba was charging.

[Conker]

The homebase is located underneath my couch so i don't pay much attention to it, but i have occasionally heard a ticking noise while the roomba was charging.

Ditto that.

[third_deg]

From 07 on IRobot switched to flame retardant grade plastics for their entire product line. This happened not because of observed issues, but as a precaution.

It's possible the CM in China changed plastic graded without approval and some units got shipped this way. The FR grade plastic has a much higher deformation temperature.

[Brett]

Regardless, there should be not observable heating, so even cr@p plastic should be fine. And why is it always the left one? Something is afoot!

(-8

[GermanRoombaUser]

[...] there should be not observable heating [...]

I do not know the specs and we all do not know the exact temperature needed to let the deformation begin. Maybe iRobot specified a certain temperature as "normal" and maybe this "normal" is above what we think should be normal. When you have current you always have thermal losses. I.e. my notebook ac/dc adapter gets "better not touch"-hot on charging.
Nevertheless you may be right.

why is it always the left one

This is THE question.
Idea: I had a look into my damaged Roomba and as you may know the inside design is not symmetrical. On the right the side brush motor is located, one would expect higher thermal losses here but it seems to be the other way round. Confusing.