roomba 550 charging error 5 (again...)

The very latest news and updates for the iRobot Roomba robotic vacuum, the Scooba Robotic Washer and the Dirt Dog workshop sweeper. All discussion and troubleshooting questions go here.

Re: roomba 550 charging error 5 (again...)

Postby pomonabill221 » July 15th, 2012, 12:55 am

Just a little tidbit of info on charging with a NON roomba supply.
IF you do not limit the current to LESS than 1.7 amps, the bot will not charge and will throw an error 2 (I believe... old age and memory), and the lady in the box will say "charging error".
I tried a bench supply set at 22.5 volts, current limit at 4 amps, and my 560 complained.
When I limited the current to about 1.3 amps, no complaining.
I slowly increased the current limit, and at about 1.7-1.8 amps, I got the error/voice warning, and the charging stopped (0 current).
SO... high current, as well as under current, does matter as well!
User avatar
pomonabill221
Robot Addict
 
Posts: 138
Joined: November 30th, 2010, 4:03 pm

Re: roomba 550 charging error 5 (again...)

Postby bhylak » July 15th, 2012, 4:24 pm

It also can do permanent damage.
Ben

Homemade Telepresence Robot:
http://mayarover.blogspot.com/
http://www.instructables.com/id/MAYA-Te ... nce-Rover/

There are only 10 types of people in the world. Those who understand binary and those who don't
User avatar
bhylak
Robot Addict
 
Posts: 1637
Joined: October 17th, 2009, 7:30 pm

Re: roomba 550 charging error 5 (again...)

Postby pfile » July 20th, 2012, 2:13 pm

Gordon wrote:ADDENDUM to my above post.

pfile, I ran across an -R3 schematic in my archives that you will need. It was created almost a year ago, explaining why I had forgotten about it! That's what happens when you get old! :-)

I'm going to link you to a post containing the d/l link to the pdf. This one: viewtopic.php?p=98692#p98692


Hi Gordon, thanks very much for the reply and the additional resources. sorry i have not checked the thread in a few days.

i have not yet mustered the desire to open up the roomba. to be honest it's limping along well enough with the new charger that i can still use it every day to clean up after the pets.

i did one more experiment this morning. i used a Kill-A-Watt to measure the power consumption during charge. the roomba pulled 26-28W AC continuously for the 30 minutes from plug-in to error 5. if we assume 80% efficiency, the battery got just about 11Wh of energy in that 30 minutes.

because error 5 apparently means the charging current was 400mA or less for 1/2 hour, it seems like a nonsensical error... because 28W at the wall is roughly 22W at the battery (again 80%) and if the battery voltage is ~20V you're talking about ~1A. it's not like the roomba is getting hot; i don't think all this energy is being dissipated as heat.

what is the actual voltage at the battery terminals during charge? i realize the answer is probably in one of the posts you linked to...

anyhow, does this indicate that the current sense circuit is bad? i know precision resistors can be fickle. are resistors R235 & R257 the current sense resistors? *EDIT YES, per schematic* perhaps just measuring the value of those resistors can be a clue here, but it's probably difficult to get a good reading of a 0.05ohm resistor!
pfile
 
Posts: 11
Joined: July 2nd, 2012, 7:25 pm

Re: roomba 550 charging error 5 (again...)

Postby Gordon » July 20th, 2012, 4:36 pm

pfile wrote:...Hi Gordon, thanks very much for the reply and the additional resources. sorry i have not checked the thread in a few days.
Welcome! And it was just by happenstance that I came across this post of yours, since I have not been receiving consistent 'new-reply' e-mail notifications from rr.
i have not yet mustered the desire to open up the roomba.
Not a problem. It can be beneficial to avoid knee-jerk operations!
... i used a Kill-A-Watt to measure the power consumption during charge. the roomba pulled 26-28W AC continuously for the 30 minutes from plug-in to error 5. if we assume 80% efficiency, the battery got just about 11Wh of energy in that 30 minutes.

because error 5 apparently means the charging current was 400mA or less for 1/2 hour, it seems like a nonsensical error... because 28W at the wall is roughly 22W at the battery (again 80%) and if the battery voltage is ~20V you're talking about ~1A. it's not like the roomba is getting hot; i don't think all this energy is being dissipated as heat.
Too much guessing by me would be required to respond, so I'd rather not. Getting direct answers by checking dc voltages throughout the charging circuit would be more satisfying.
what is the actual voltage at the battery terminals during charge? i realize the answer is probably in one of the posts you linked to...
Your Fast Charger's output characteristic of CURRENT ROLL-OFF curve could tell what that battery's terminal voltage is during normal (when charging FETs are held in the continuous conducting state) high-rate charging -- if you had your own curve. Until you plot that curve we will have to be satisfied by looking at this old one:Image
This curve suggests that at the start of re-charging a well depleted battery the voltage applied to the battery (which you could measure between main_PCA terminals "VBAT" & "GND") might be in the 10-15 volts range (the PSU's current-control circuit will have caused the PSU's output voltage to fall in order to hold output current in check).
anyhow, does this indicate that the current sense circuit is bad? ... it's probably difficult to get a good reading of a 0.05ohm resistor!
While such resistance shift is always possible, I do think you will find the shunt to be OK.

My DMMs, and a 0 to 99ohms ESR meter have provided reasonable resistance checks on Roomba's master shunts -- not that I can claim better accuracy of any of those meters than the mfrg accuracy for those 0.1 ohm resistors!

One of the better ways to determine the shunt's resistance is to pass one to two amps of measured current through it in a bench test setup. With a specific current injected, just measure voltage drop to ensure dV is in the range 0.05V to 0.1V.
Gordon
Robot Master
 
Posts: 4129
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: roomba 550 charging error 5 (again...)

Postby pfile » August 13th, 2012, 2:26 pm

Gordon wrote:One of the better ways to determine the shunt's resistance is to pass one to two amps of measured current through it in a bench test setup. With a specific current injected, just measure voltage drop to ensure dV is in the range 0.05V to 0.1V.


yeah, i'm going to start here. i have a bench power supply so i'll test the current shunts. i've been out of the country for 2 weeks and have not had time to work on the roomba. one interesting thing is this: the roomba battery was completely dead when i returned. i plugged it in and got a 'fast blink' orange charging light. i then reset the roomba and plugged it in again, and it started charging. it has not yet thrown the error 5 after a little over 2 hours. the wattsup says it's consumed about 60Wh, and is drawing 29W continuously from the wall.

anyway something must be wrong in the current measurement circuit somewhere, because when it throws error 5 the WattsUp indicates that the DC current had to have been >1A for the entire time.

edit: the roomba charged fine from the dead battery condition. for the next cycle, instead of topping it off after it's mission, i let it run all the way down again. it again charged for about 3 hours with no error 5, and the watts up indicated that it consumed about 70 watt-hours.

i'm left wondering if i have simply been trying to charge the roomba when it was already almost full. of course the charging circuit should be able to handle this condition without error...
pfile
 
Posts: 11
Joined: July 2nd, 2012, 7:25 pm

Re: roomba 550 charging error 5 (again...)

Postby wrattspider » August 26th, 2012, 6:26 pm

This thread has been very informative. I have had a similar problem on my old Roomba 530 (charging error 5 every time I try to charge). I bought a replacement battery, but still get the same issue, regardless of whether directly charging or through the home base.

Tonight I tried charging directly from the charger for 15 minutes, while measuring voltage at the battery terminals. What I see is that the terminal voltage stays at approximately 17.3V the whole time, then when I cut power at the end of the charge it drops to 16.3V.

I'm assuming this means that the charger is faulty (I measure 22.5V with the charger unloaded, and I am expecting near 22.5V under load). Would this be a reasonable assumption?

Sorry if this has already been covered, but I haven't seen any documented cases where the power charger was tested and found to be faulty.
wrattspider
 
Posts: 4
Joined: August 26th, 2012, 6:05 pm

Re: roomba 550 charging error 5 (again...)

Postby TechGuy » August 26th, 2012, 7:37 pm

No load voltage is 22.5V. Under load is around 17 -18V. You power supply is working.

After charging the battery for 15 minutes, How long the battery last?
Charging battery directly: 400 Series, 500 Series,. How to Desolder
For Sale: Black CHM 2 Ball Bearings set
User avatar
TechGuy
Robot Addict
 
Posts: 1201
Joined: February 17th, 2009, 11:52 pm

Re: roomba 550 charging error 5 (again...)

Postby Gordon » August 26th, 2012, 8:00 pm

wrattspider wrote:... I tried charging directly from the charger for 15 minutes, while measuring voltage at the battery terminals. What I see is that the terminal voltage stays at approximately 17.3V the whole time, then when I cut power at the end of the charge it drops to 16.3V.
I don't see that as surprising. If you were to repeat what you just did, but measure PSU-output charging current, I expect you would measure more than 1.3 amperes dc current.
Did you study the R2 PSU's E vs I plot a few posts above yours?

If you were to carry on that 'direct' charging for half an hour, you ought to sense the battery heating (BTW, if you do such a long period with direct PSU to battery connection, you should periodically touch the battery all the while and not waiting for 30-mins to pass! If you feel warmth on its exterior it is a lot hotter inside the cells. Better yet would be to measure battery-thermistor resistance and stop charging current if resistance falls below about 5000 ohms.
...Sorry if this has already been covered, but I haven't seen any documented cases where the power charger was tested and found to be faulty.
I can't recall any either.
Gordon
Robot Master
 
Posts: 4129
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: roomba 550 charging error 5 (again...)

Postby wrattspider » August 26th, 2012, 10:02 pm

Wow... that's quick response! 2 replies within 1.5 hours!

The previous test I did with my old battery (which might be bad). I just repeated the direct charge on my new battery. At the end of the 15 minute charge I checked voltage at the battery terminals with and without the charger plugged in, then lifted one of the leads and measured charging current (therefore not simultaneous with voltage). I measured 17.08V charging, 16.88 V once the charger was unplugged, and 1.42A (wow, surprised me there!).

It ran for more than 10 minutes (then started getting alternating "error 5, spin left/right wheel" errors). So I tried charging it again. After 30 minutes, the charging error 5 popped up.

From this, I now assume I've got a motherboard problem, and the battery and charger seem fine.
wrattspider
 
Posts: 4
Joined: August 26th, 2012, 6:05 pm

Re: roomba 550 charging error 5 (again...)

Postby Gordon » August 26th, 2012, 11:33 pm

wrattspider wrote:Wow... that's quick response! 2 replies within 1.5 hours!
Some of us are interested in the odd cause(s) of err5.
... I measured 17.08V charging, 16.88 V once the charger was unplugged, and 1.42A (wow, surprised me there!).
No surprise to me!
It ran for more than 10 minutes (then started getting alternating "error 5, spin left/right wheel" errors).
Please take that sentence apart and reconstruct it. As it is, I make little sense out of it. Why are wheel behaviors being associated with err5?
So I tried charging it again. After 30 minutes, the charging error 5 popped up.

From this, I now assume I've got a motherboard problem, and the battery and charger seem fine.
I don't see why the battery is in the clear; and, don't you think the PSU should be load tested (at a load current in excess of 400mA) for a longer period than 30-minutes?
Gordon
Robot Master
 
Posts: 4129
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: roomba 550 charging error 5 (again...)

Postby wrattspider » August 27th, 2012, 12:29 am

'Please take that sentence apart and reconstruct it. As it is, I make little sense out of it. Why are wheel behaviors being associated with err5?

I'm not sure whether there is a relationship, it just seems that the wheel errors ("Uh-oh" code 5) and the charging code 5 errors started happening at the same time.

I don't see why the battery is in the clear; and, don't you think the PSU should be load tested (at a load current in excess of 400mA) for a longer period than 30-minutes?


If I had the equipment to do a good load test (maybe just a power resistor?) it would be a good idea. I think the battery is in the clear because it's been charging up to 16V+ and it is rated for 14.4V. Also, I have been testing an old battery and a new battery and neither fixes the problem.

I just took the Roomba apart to get a good look at the PCB. All I noticed was some pet hair that had gotten through the center wheel well and was on the PCB. I cleaned it and reassembled and will retest tomorrow.
wrattspider
 
Posts: 4
Joined: August 26th, 2012, 6:05 pm

Re: roomba 550 charging error 5 (again...)

Postby Gordon » August 27th, 2012, 1:14 am

wrattspider wrote:
'Please take that sentence apart and reconstruct it. As it is, I make little sense out of it. Why are wheel behaviors being associated with err5?

I'm not sure whether there is a relationship, it just seems that the wheel errors ("Uh-oh" code 5) and the charging code 5 errors started happening at the same time.
If Roomba issues Uh Oh + five beeps while performing in Clean Mode, that is 'normal' fault behavior. But, if Roomba also issues charging-err5 (Check Robot = red + five blinks) while in Clean Mode, that is abnormal and suggests the MCU is sick.

Charging err5 is associated with "Charging Mode". A correctly operating Roomba has no way to make a drive wheel rotate while the MCU is engaged with the battery charging task.
I don't see why the battery is in the clear; and, don't you think the PSU should be load tested (at a load current in excess of 400mA) for a longer period than 30-minutes?
If I had the equipment to do a good load test (maybe just a power resistor?) it would be a good idea.
To load the charging PSU, a 20 ohms resistor with 50 watts power rating would do it, but, that's not likely something you'd have around the house. An alternative for that task would be an incandescent lamp filament, actually a pair of automotive lamp filaments connected in series. The T-211 interior lamp (looks like a glass cartridge fuse but larger) draws about one amp when powered with 12V. Two in series would give a 24V x 1A load. But, powering it with the 22.5V PSU would run current through the pair at somewhat less than an amp, but safely more than the 400mA err5 level.

If interested, you can buy the T-211 lamps, and four small-size battery-clamps to grip their contacts, at you local hardware store.
I think the battery is in the clear because it's been charging up to 16V+ and it is rated for 14.4V. Also, I have been testing an old battery and a new battery and neither fixes the problem. ...
I tend to agree that the battery is not at fault, but I want to point out two things: 1) voltage measurements say little about a battery's capability to perform as expected, and 2) "a new battery" does not always equal a battery capable of delivering its rated capacity.
Gordon
Robot Master
 
Posts: 4129
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Re: roomba 550 charging error 5 (again...)

Postby pomonabill221 » August 31st, 2012, 8:15 pm

Maybe try placing the load resistor/lamps across the battery while measuring the voltage to see if the battery droops when loaded, to see if the battery does have some capacity.
User avatar
pomonabill221
Robot Addict
 
Posts: 138
Joined: November 30th, 2010, 4:03 pm

Re: roomba 550 charging error 5 (again...)

Postby pfile » September 2nd, 2012, 7:39 pm

i guess the time has finally come for me to open up this roomba. after that deep discharge, it gave me about 2 or 3 error 5 free charging episodes, but has since gone right back to err5 after 30 minutes, regardless of the state of charge at the beginning of charging. if i reset the roomba and plug in again, it charges for another 30 minutes... repeating that a few times it eventually ends the charging without error 5.
pfile
 
Posts: 11
Joined: July 2nd, 2012, 7:25 pm

Re: roomba 550 charging error 5 (again...)

Postby wrattspider » September 17th, 2012, 12:31 pm

Sorry for the long delay in responding. I have done another test which (I believe) points to something other than the battery or the power supply as the source of my charging error 5. I had to order a 2nd meter and some alligator clips so I could get voltage and current simultaneously.

First I flipped over the roomba and took off the bottom cover. I have taken the alligator clips, and clipped the battery terminals to the Roomba's charging tabs, with one multimeter in series (to measure current) and another in parallel (to measure voltage). I have also did extension clips to keep the battery's thermistor in play. Then I plugged the power supply directly into the Roomba's side charging port.

What I measure when I do this is 15.7V at 0.28A continuously (for a couple of minutes). My understanding is that the charging error 5 will happen when the charging current is below 0.4A for more than 30 minutes. In this test, the current never exceeded 0.4A, so the error will be declared 30 minutes after plugging in the power supply. (I didn't wait the 30 minutes, as I had to hold several of the terminals in place).

If I then direct-wire from the power supply to the battery, the 15 minute test starts at 16.7V, 1.4A and ends at 17.8V, 1.39A.

Then I ran another test where I plugged into the home base, put the battery back in the roomba, and clipped between the home base terminals and the terminals on the underside of the roomba. Got a voltage between 20.8 and 21.4V (bounced around every few seconds), and a current between 0.28A and 0.34A. After 30 minutes, the charging error 5 was declared and the levels dropped to ~22V, 0.08A.

So I conclude that, at the very least, there is a problem somewhere between the power supply plugin and the roomba battery terminals. Is there something I can try unplugging on the system card to find the cause? Thanks!
wrattspider
 
Posts: 4
Joined: August 26th, 2012, 6:05 pm

Re: roomba 550 charging error 5 (again...)

Postby TechGuy » September 17th, 2012, 2:06 pm

My 560 has similar problem except it charge with 0.00A current. I was planning to trace the MOSFET gate driver circuit last weekend. But I got side tracked without doing anything.
viewtopic.php?f=1&t=15704
Charging battery directly: 400 Series, 500 Series,. How to Desolder
For Sale: Black CHM 2 Ball Bearings set
User avatar
TechGuy
Robot Addict
 
Posts: 1201
Joined: February 17th, 2009, 11:52 pm

Re: roomba 550 charging error 5 (again...)

Postby Gordon » September 18th, 2012, 2:22 am

wrattspider wrote:... I flipped over the roomba and took off the bottom cover. I have taken the alligator clips, and clipped the battery terminals to the Roomba's charging tabs, with one multimeter in series (to measure current) and another in parallel (to measure voltage). I have also did extension clips to keep the battery's thermistor in play. Then I plugged the power supply directly into the Roomba's side charging port.

What I measure when I do this is 15.7V at 0.28A continuously (for a couple of minutes). My understanding is that the charging error 5 will happen when the charging current is below 0.4A for more than 30 minutes. In this test, the current never exceeded 0.4A, so the error will be declared 30 minutes after plugging in the power supply. ...
You are off to a good start, wrattspider! But one thing to keep in mind is (as best I know) err5 only relates to the high-rate period of battery charging -- which is the initial and long term portion of charging.

To be certain that Roomba has locked in to the high-rate (~1.3A) session, either charging current must be confirmed (which your data did not) or one must verify that the charging FETs have their gates continuously pulled to a low voltage (by the MCU).

In one of my replies to member pfile (previous page) I listed all the components across which to measure voltage drops, or other voltage states. I think you could use that list, see it via this link: viewtopic.php?p=109257#p109257
In that same post I provided a link to a Power-Forms diagram that shows all those listed elements hooked together. As a teaser, here is a screen-shot of the upper left corner of that drawing:
UpperLeftCornerR3PFs.jpg
{Click image to enlarge}
If you start at the left side of that drawing, at either of the two charging inputs, you can follow conventional current through the listed elements half way across the diagram, and then turn downward through the charging FETs and into the battery.
You will have to d/l that pdf file to view the entire charging current path, and return of current to the PSU.

To follow through on those measurements it will be necessary for you to obtain access to the upper face of your Roomba's main_PCA.
...If I then direct-wire from the power supply to the battery, the 15 minute test starts at 16.7V, 1.4A and ends at 17.8V, 1.39A. ...
Well, that confirms the PSU is doing its job -- at least for 15-mins.
...Is there something I can try unplugging on the system card to find the cause? Thanks!
I can't think of any.
Gordon
Robot Master
 
Posts: 4129
Joined: April 6th, 2005, 2:02 am
Location: Santa Ynez, CA USA

Previous

Return to iRobot Roomba and Scooba Chat

Who is online

Users browsing this forum: ChristianNO, Fraggboy, instrukcije, joanquro, magura, Neato Robotics, TechGuy, viper001 and 208 guests