Battery issue 0004. Please press okay to restart. SOLVED

[mmm2000]

My Neato XV14 is about 3 years old or so. I have 3.4 software on it after update. I changed the batteries about a year ago and had no issues until about a month back when while charging I got the battery issue 0004 error.
After several reboots the robot would charge and work fine. Then on the next cleaning it would have the error again. Same thing, a few reboots and then it would charge and work.

After about one week it just won't charge the batteries anymore. Contacted Neato and after a few things they had me do and said after all to replace the battery. The conversation with Neato support is at the bottom of this post.
Before the robot died altogether I measured the voltage of the batteries and it was around 7V each. After trying to charge the robot and several other reboots the batteries got depleted and one was at 2 volts and the other less then 1V.

I ordered another set of batteries from Amazon, Li-Ion this time. The robot starts without a glitch, I do the "new battery" in the menu and it says batteries are low, needs recharging. I let it charge and after about 10 mins or so the dreaded battery issue 0004 come back on... With 2 brand new batteries. After a few reboots I measured the voltage in the batteries. 7.1V and 6.9V. The docking station had about 24V without the robot and it had about 13.3 with the robot charging.
After a few more recharge tries, the robot dies. I pull the batteries out and I get 2.2V on one and less than 1V for the other.

Before it died I was measuring the voltage of the docking station. With the robot docked and charging, the voltage was dropping steadily but very slowly. From 13.3v it dropped to about 12.6, about 8 minutes later, right before the robot died. After it died the docking had an output of only about 2V. As soon as I removed the robot the voltage jumped to 24V. But now the robot is dead, most likely because the batteries got drained.

My conclusion is that the batteries were not the problem here at all. The older batteries lasted as long as when new, when the battery issue 0004 appeared the first time, a month ago.

Why did the charging station voltage drop to 2V when the robot died? I also turned off the dust bin On/Off switch and the voltage remained at 2V, while robot was docked.

Has anybody had success in fixing the 0004 battery issue? Any advice? Now I am stuck with the new, very expensive batteries too...


*****************************************************************************************************************

" Thank you for contacting Neato Customer Care. My name is Michiko, and I will be happy to assist you today! *********, yes, most likely your Neato just need to have new battery. I have also contacted my supervisor on this matter, and they also have told me to suggest you to purchase new battery. Neato battery last about 2-3 years, so your Neato battery is due date for expire.

Please follow the steps that I have send to you on how-to when you get the new battery.

Please do not hesitate to contact me if you have any concerns or questions. "


*********************************************

" Thank you for contacting Neato Customer Care. My name is Michiko, and I will be happy to assist you today! *******, I understand your Neato is having battery issue 0004.

I would be more than happy to assist you with this, what I would like to do is walk through a few troubleshooting steps with you.

What causes this error: The Neato has been away from the charge base for too long, and is low on current or voltage.

How to resolve this error:

Check the charge base to make sure all connections are secure. On an XV base, open the base and check internal connections as well.

Fully charge the Neato for at least five hours.

Hard reset by unplugging one battery OR turning the power switch off for 3-5 minutes. (Note: If the customer has a power switch, make sure to always use the switch to avoid unnecessary strain on the batteries)

Turn Neato back on, place on the base.

If this resolves the Battery Issue,

Navigate to Menu > Support > New Battery. Neato will say "Thank You." Press the Back arrow twice to get back to the main screen.

Fully charge and run for 3 complete charging and discharging cycles.

If you have completed all of these troubleshooting steps and your issue persists, please reply via e-mail or call us at 877-296-3286 for further assistance."

[glnc222]

My guess is the system board has a defect in the charging and power distribution components, and needs entire replacement.
If you can see a message on the LCD while the voltage on the docking plates is 2v, the computer must be running and the only power source could be the battery. Something might be shorting out the docking plate power inside the Neato and there can be some protection circuit in the dock shutting it down. With defects the detection of the battery and dock can be messed up some way the system reports a battery error of some kind.
Just a guess. The normal way to test this possibility is to replace the system board and see whether that solves the problem. There is no inexpensive way to repair defective computer systems except in certain cases.
Unless one of the engineers has seen this before and knows a particular component to replace, that is what you have.

[Lewiy]

This issue may be caused by malfunction of the protection board, installed in the charging base. The diode D1 is often referred to as the culprit in Russian forum. But other causes are also possible.

[glnc222]

Agree with Lewiy. There have been other reports of defective charging docks requiring replacement. The unit needs to be tested on another dock before replacing the system board.
The short circuit protection is also inside the power brick inside the dock, and these have needed replacement.
The design of the dock changed a few years ago and the dock and brick must be bought together (old ones have 2.49 amp rated brick, newer ones smaller; old docks cannot charge newer vacuums, an LCD message will advise).

[mmm2000]

My guess is the system board has a defect in the charging and power distribution components, and needs entire replacement.
If you can see a message on the LCD while the voltage on the docking plates is 2v, the computer must be running and the only power source could be the battery. Something might be shorting out the docking plate power inside the Neato and there can be some protection circuit in the dock shutting it down. With defects the detection of the battery and dock can be messed up some way the system reports a battery error of some kind.
Just a guess. The normal way to test this possibility is to replace the system board and see whether that solves the problem. There is no inexpensive way to repair defective computer systems except in certain cases.
Unless one of the engineers has seen this before and knows a particular component to replace, that is what you have.

Actually the robot was dead, when I measured 2.2V while docked. Nothing on the LCD screen. When the LCD screen was on, before it shut off, the voltage was 12.6V or so. As soon as the robot shut off, the same second the voltage dropped at the docking station. I was measuring while that happened.

What is strange is that even with no batteries in the robot, if I dock it, the voltage at the docking station measures 2.2V. I think you need good batteries with voltage above 6.5V in order for the computer to boot up. So now I can't even test anything else because the batteries are discharged.

Can anybody test the voltage at the docking station with the batteries removed on a working robot, while the robot is docked? Or does anybody know what the voltage is supposed to be with or without batteries?

How can I manually charge the batteries, just to get enough juice to get it started and maybe hook it up to my PC to do some more diagnostics? Can I do it with the neato charger and connected directly to the 2 batteries?

My charger uses 120V.

[glnc222]

My experience is with the older system board, with the side charging plug, replaced a few years ago. On this board the unit would operate on the dock without the batteries installed, issuing an urgent "critical battery failure" message and showing red ring light, not just orange. I suspect the same thing applies to the newer system board, because some failure of the battery is a normally possible condition to plan for.
You can also find sometimes that the USB connection is operating even if the LCD is not working.
This is one way to test whether the battery is a problem compared to a broken cpu, because you could run the computer on the dock power to show at least part of it is working. The battery power and charging circuits can be separately damaged (I have damaged them myself in experiments with alternative batteries and what not).
There is only so much that can be done without a second unit to swap docks and parts to see which one is faulty.

There are external chargers which can charge the batteries to test them outside the unit.

There is a circuit board inside the dock which works in combination with the system board for charging making it difficult to test the dock without a working Neato, without more detailed knowledge of the circuits.
The older Neato's had a small resistor in the back across the charging contacts to load the dock as a signal the bot was connected so charging could be enabled. This resistor has been eliminated in more recent designs, more efficient. However, the newer docks will charge the older vacuums. So perhaps a dock could be tested by connecting such a resistor across the contact plates and seeing if the voltage stays up. I think around 360 ohms 2 watt resistor, dissipating one and a half a watt at a 24v dock voltage. Gets hot is one reason eliminated.
[edit] Three very common 1K ohm 1 watt resistors in parallel would be similar to a 2 watt resistor.

[Lewiy]

Try connecting the power brick directly to the contact pads to bypass the faulty protection board in the charging base. If the robot turns on and starts charging after that, you will know exactly where the issue nests.

[mmm2000]

Try connecting the power brick directly to the contact pads to bypass the faulty protection board in the charging base. If the robot turns on and starts charging after that, you will know exactly where the issue nests.

I did connect the charger directly to the robot, bypassing the circuit in the charging base. I had to improvise some wires to do that. I used copper speaker wire of about 3 feet length. I think is 14 gauge or maybe 16.

The robot seemed to be charging for about an hour. I was all happy it did not give me the 0004 error. But after that I got the battery issue 8006 error. No more 0004. The 8006 from what I read so far it means it takes too long to charge the batteries. Mind you, the light was still orange even after about an hour of charging. From what I remember the light would turn green after about 45 mins or so of charging when the robot worked fine.

Also when I measure the voltage, when I bypassed the charging base, I got a voltage jumping up and down from 10V to 14V every second or so. Is that even normal? This is with the robot charging. Without the robot, the charger gives a steady 24V. When I use the charging base and its circuit board for charging I get about 14V steady while charging the robot, but then after 10 mins or so I still get the 0004 error.

Now I reinforced the connections with a few loops around the charger's tip and the robot's back plates. Maybe the connection was not very sturdy and that is why there was not enough current flowing to the batteries in whatever time Neato expects it to flow. I will post an update soon. It has been charging for 20 minutes now.

I am not sure where the problem is here. Any advice? Are those voltage readings normal? Did anybody measured them on a good working robot/charger?

The side plug seems to not charge at all. I think it was deactivated on the newer Neatos.

[glnc222]

The circuit board inside the dock is part of the charging circuit. This circuit converts the 22-24v dock voltage to a continuously varied voltage at the battery terminals as charge is accumulated. The battery terminals will show a voltage near 18.5v at the end of charging and the battery is full.
I do not know what the charging circuit would do without the components inside the dock.
The fluctuations observed could be due to defects in either the circuit in the dock or in the power brick.
You must use the correct size of power brick matching the model of the dock.
I suspect the power brick is bad. These have failed in the past.

The power brick should deliver around 22v at two full amps through a a couple of 12v 25 watt lamp bulbs in series, or a bunch of ten watt resistors in parallel. It is possible heavy load is shutting it off.

[mmm2000]

The circuit board inside the dock is part of the charging circuit. This circuit converts the 22-24v dock voltage to a continuously varied voltage at the battery terminals as charge is accumulated. The battery terminals will show a voltage near 18.5v at the end of charging and the battery is full.
I do not know what the charging circuit would do without the components inside the dock.
The fluctuations observed could be due to defects in either the circuit in the dock or in the power brick.
You must use the correct size of power brick matching the model of the dock.
I suspect the power brick is bad. These have failed in the past.

The power brick should deliver around 22v at two full amps through a a couple of 12v 25 watt lamp bulbs in series, or a bunch of ten watt resistors in parallel. It is possible heavy load is shutting it off.

Update: So I have been charging the robot like I said directly from the brick and so far has been charging successfully. Now the light is blinking green, sign that the batteries are mostly charged.
The only concern is that is has been charging for about 6 hours and it has not stopped yet. It stayed on the orange light for about 4 hours or so.

But this is with the new Li-Ion batteries I purchased the other week and have never been fully charged. I think now that the robot is fine and my problem lies with the charging dock or with the power brick or both.

From what "glnc222" said I understand that I have to test the brick with a 50 watt/12 volt load. Can I use a car light bulb rated at 55W for this? So I hook up the light bulb and then measure the voltage of the brick with the bulb connected and I should have about 22V. Am I getting this right?

[glnc222]

Be careful charging those lithium ion batteries (with internal adapter to NiMh charging) with a defective power brick.
The lithium batteries stop charging when the voltage applied to the battery terminals reaches 16.8v, at which point an artificial high temperature signal is sent to the Neato NiMh charger to stop it. The charging may be continuing because the brick is not supplying enough voltage -- not really charging.
Don't mess with lithium batteries. There are all sorts of protections built into them, though, so safe.

When the Neato internal charging circuits go into the blinking green phase, the charger stops when rapid temperature rise is sensed from the thermistors. You should look at the charging current reported over USB, with Neato Control Program or terminal emulator (the Neato driver must be installed in the PC). It might show no current going into the lithium batteries, waiting on a higher voltage. The circuit is probably not equipped to detect defects in the charger and signal an error. There could also be some role of the dock circuit you have bypassed.
Over USB you can also see the internal Neato parameter "per cent fuel gauge" showing the estimated degree of charge on the battery. The parameter can be manually set when incorrect. You can also observe the thermistor temperature readings (only the high side pack temperature, with 14-16v red wire, controls the charging).
Also set New Battery on LCD to reset. Disconnecting the battery will supposedly recalibrate this as well, with a mandatory recharge cycle.

The USB plug is under a little rubber cover on the side.

Failure of the charger to apparently stop, seemingly locked in charging mode (even if not actually charging) is associated with a defect in diode D14 on the system board, from heat. Separate thread on this. Many repairs done around the world.
http://www.robotreviews.com/chat/viewto ... 52#p126652

[mmm2000]

I took the readings with the Neato Control Center. Do these look ok? Some of the sensors are disconnected. The robot is in pieces on my floor for now, until I figure out how to fix it. Also the ChargeVoltInmV=11356 changes 2 or 3 times per second, from 11V to 17V. This is I assume because I am charging just with the brick and not using the dock.

____ Discrete: ____=
SNSR_DC_JACK_CONNECT=False
SNSR_DUSTBIN_IS_IN=True
SNSR_LEFT_WHEEL_EXTENDED=False
SNSR_RIGHT_WHEEL_EXTENDED=False
LSIDEBIT=False
LFRONTBIT=False
RSIDEBIT=False
RFRONTBIT=False
____ Analog: ____=
WallSensorInMM=OFF: 60
BatteryVoltageInmV=15276
LeftDropInMM=OFF: 0
RightDropInMM=OFF: 0
LeftMagSensor=OFF: 32768
RightMagSensor=OFF: 32768
UIButtonInmV=3308
VacuumCurrentInmA=0
ChargeVoltInmV=11356
BatteryTemp0InC=23
BatteryTemp1InC=23
CurrentInmA=182
SideBrushCurrentInmA=0
VoltageReferenceInmV=1224
AccelXInmG=-80
AccelYInmG=-16
AccelZInmG=1004
____ Buttons: ____= (only in test mode)
BTN_SOFT_KEY=OFF: False
BTN_SCROLL_UP=OFF: False
BTN_START=OFF: False
BTN_BACK=OFF: False
BTN_SCROLL_DOWN=OFF: False
____ Charger: ____=
FuelPercent=64
BatteryOverTemp=0
ChargingActive=1
ChargingEnabled=1
ConfidentOnFuel=0
OnReservedFuel=0
EmptyFuel=0
BatteryFailure=0
ExtPwrPresent=1
ThermistorPresent[0]=1
ThermistorPresent[1]=1
BattTempCAvg[0]=23
BattTempCAvg[1]=23
VBattV=15.26
VExtV=13.15
Charger_mAH=0

[glnc222]

All it says is that the charging is not working. The battery voltage is low, the fuel gauge is low, the battery is not charged, the voltage is oscillating abnormally. The battery is not charging as the current is 182ma about what is used to run the cpu. I think it goes heavy negative 2 amps when charging.
All I know is try a new system board and dock.
Fixing diode D14 is inexpensive, can't hurt. You need to test the power brick etc. No magic wands here. It is not much different than malfunctions in a PC -- lot of parts.

[edit] You might try manually setting the Fuel Gauge to 100 per cent, see what happens. But I don't like the battery terminal voltage, looks less than full charge.

[Lewiy]

It may be that your li-ion batteries irreversibly damaged after a deep discharge. Try to repeat the tests with NIMH batteries. Make sure all the contacts in the charging circuit are reliable. You could write the charging log by repeating the command GetCharger every second. Analysis of the log would allow understand many things. If you have a laptop power brick with suitable characteristics (voltage 18-19 V, current 1-1,5A), then you can try to charge with it.

[glnc222]

To analyze problems with the Neato system board and dock circuits and power brick, the regular NiMh battery is better.

The commercial AnewPow/LithiumPowerInc lithium batteries are protected against damaging deep discharge by the included circuit board. Home made batteries only will be subject to this damage, if they lack protection boards.

When the AnewPow battery is disconnected for a long time it may go into a "sleep mode" to turn off the circuit board.
The battery will appear dead. The cells are disconnected.
This prevents any discharge while in storage. The lithium cells do not self-discharge at any significant rate.
To wake up the circuit board a 9v battery can be applied to each pack separately, outside the neato, for thirty seconds.
I think you will then see the full battery voltage on the battery terminals with a multimeter, outside the Neato.
Neato does not apply the charging voltage until the battery is detected, so it will not wake up the circuit. A missing-battery error will be reported on LCD.

You can charge the AnewPow battery with external chargers for the NiMh packs. Thermistor control is needed.

Also if the lithium cells are discharged too far by missuse, the protection circuit board will disconnect the cells to prevent damage. To activate the battery some charge must be supplied by an external charger. The charger must regulate the charging voltage to the proper level for charging. Connecting a power brick directly to the battery can supply an excessive voltage. The maximum charging current can be exceeded. Excess voltage will trip the over-voltage protection circuit and disconnect the cells.
You need proper charging equipment.

You cannot tell anything without a working dock and system board.

[mmm2000]

I charged the battery to 99% using my laptop charger (19v and 3.4A), bypassing the neato dock. At the end of the charging it game me the 8006 error. The charging took all day using 2 different bricks. For the first 70% I used the neato brick that took very long. Then I used the laptop brick and it finished the charging much quicker. About 45 mins it reached 99%, before the 8006 error. After restart the battery level said to be at 19% again. Then I tried hooking up the laptop brick to the neato dock and I tried charging it like that. Like before after about 10 mins I get the 0004 error. This makes me conclude that the neato circuit dock is at fault, not the brick.

So far I can say that the robot needs its dock for a correct charging cycle. Just the brick is not going to do it.

Either way I decided to try another charging dock. Just placed an order on ebay. 910-0075 and 24V with 1.67A. Exactly like my own dock, except mine is 910-0089. I could not figure out what the difference was, but both versions are for XV14 and others. This is the last thing I want to try before I trash the robot. Not willing to spend anymore time and $ on a new main circuit board. Will receive it before Friday. I will post an update after trying it. If it doesn't work, I will return the parts I just got (batteries and dock, if possible) and maybe buy another Neato, D75 most likely, but this time with extra 3 years of warranty. If I get 4 years out of it I think I will be happy.

My current Neato lasted just a little over 3 years. I found some dates on some of the parts inside, as April 2012. I think I must have purchased the robot summer 2012.

[glnc222]

The 1.67 amp docks are the newer ones required for the newer system boards, but also charge the older bots. The older docks were 2.49 amps and are rejected by newer system boards (now some years old). Also with the newer docks, more components are moved from inside the bot into the dock, an extra inductor or something. Basically a high frequency switching power supply with adjustable voltage, current regulation, constitutes the charger. The brick supplies a fixed voltage, the charger supplies a fixed current.

[Medtech]

In my opinion you should sell the robot as it is difficult to shift and repairs to another owner, perhaps the next owner will be more literate in matters of repair of electronic parts.

[mmm2000]

SOLVED !

Thank you all for the replies.

Today I received the new "used" base I purchased on Ebay for $25. Hooked it up, then about 30 minutes later the battery was fully charged with no issues whatsoever!

I took the robot upstairs and released it for a cleaning run. 2 bedrooms, one hallway and 2 bathrooms. Went out for some grocery shopping and came back to find everything nice and clean and the robot was already fully charged and with a dust bin quite full.

Conclusion to this saga:

1. Battery error 0004 has nothing to do with the battery. It is 100% the charger (base or brick not sure, will need to test further). So if you get this error, first thing to replace is the charging base. My suspicion is that the circuit board in the charging base (yes, there is a small one in there) went bad. I inspected the board and there was no obvious sign of anything being burned or any kind of damage. It would be curious to know which component on that board failed. At least in my case this was the fix. And now I got 2 sets of batteries. My original batteries that had nothing wrong with them and thanks to Neato's customer support a new set (Li-Ion) of batteries, because they said to replace the batteries.

2. Don't waste any time with Neato customer support. They are useless, at least from my experience.

3. If you ever buy a robot vacuum, spend a few extra bucks and get the extended warranty if available. In US you can buy 3 extra years of warranty for about $32 or so. That is 4 years of total warranty. Peace of mind.

4. These forums have the best solutions to the robot problems. Thanks to whoever runs this site! These robots are very expensive and few people are willing to toss it out whenever there is a problem.

[vic7767]

NeatoRobotics does not offer LI-ion packs, can you provide pictures of the packs sent from NeatoRobotics?