Help a noob?? 18v NiMh vacuum cleaner

[Nick H]

Hello everyone,

this is my first post here, very glad to have found what seems to be a helpful and caring community for electronics.

I'll try and stay on point: I need help in my second project, which is repairing an 18v vacuum cleaner.

https://www.pccomponentes.com/rowenta-air-force-extreme-18v-aspirador-escoba-sin-cable

What's wrong with it?? Almost certainly just the dead battery, and no way to charge it. (A light comes on dimly when switch is moved into "turbo" position)

Why not just buy a charger?? It's 25€ which I don't want to spend without knowing if it is worth repairing/modding (was thinking of trying to upgrade it to Li-ion if it all works as the unit actually got really good reviews when it was released and wasn't particularly cheap.)

So here is my thought process and where I need help: I'd like to try and charge the battery (which has been dead for about two years, its NiMH, sitting at 10v) by whatever means possible, and then test the hoover to make sure the rest of it works as it should, before committing money to it.
As I understand, some people have had good success reviving NiMh packs using very slow charging and doing a few full cycles - that would be the most cost effective and simple solution.

The problem is [almost entirely due to my noob ignorance for sure] - the hoover is marketed as 18v. However the battery pack is 20s. I'm assuming this as no other configuration makes sense, there are definitely twenty cells, and NiMh has a nominal voltage of 1.2v and a max of about 1.4v, per cell. There is no data sheet online for this particular pack, and the only marking is a sticker which says 18v 2000mah (again making sense of the 20s configuration assumption with the 2000mah rating).
https://www.electrotodo.es/bateria-aspirador-air-force-extreme-18v-rh5233    - the replacement part.
So assuming my assumption is correct, the max voltage for the pack should be 28v, and nominal 24v.

To make matters more confusing [for me at least], the replacement charger is 36v, 250mA. Now this does actually make some sense to me, since the only PCB in the hoover was clearly designed for 6 different configurations, 12v, 18v, or 24v NiMh, or 14.4v, 18v, or 25.2v Li-ion. (here is the 24v version https://www.electrotodo.es/circuito-aspirador-air-force-24v-rowenta-rh5313) so I guess it just steps it down as per each configuration.

Now I don't have a 36v adaptor to plug into the hoover board, nor a 28v to go directly into the battery, but I do have a 24v AC/DC adaptor. The adaptor I have spits out 600mA and actually measures 23.5v on my [poor quality] multimeter. I figured that I could use it under observation to get the batteries somewhere near 24v in short goes (as really most manufacturers seem to recommend 0.1C long charging as the easiest for NiMh) but it got pretty hot pretty quick so I haven't tried again.

Now that adaptor is not usually meant to charge a battery, it usually just runs a device which is on or off - so I figured [afterwards] that it would be missing a diode (or some more advanced circuit) to stop reverse charging?? Could I put one on one of the output terminals to avoid this? Is it that simple?

Also would I be OK to put a resistor also in series, to get that current down from 600mA to 100mA (which I figured would be perfect for a very slow charge to try and revive it)?? I'm hoping my awful maths skills are OK for the calculation haha, 24v/0.1A = 240ohms resistor?? I'm a bit embarrassed to say that I don't feel confident in applying ohms law, since I'm not sure if I'm putting in the right values!! My logic is that I want 0.1A from the 24v supply so that's what I calculated. If I got this wrong and there is an easy way to correct me, it'd be greatly appreciated!!

So I guess that's my questions for now. Sorry for the super long post, I'm trying to be as clear as possible. Thanks in advance for any help - there will undoubtedly be more questions in the future. And any suggestions/tips for changes of direction/learning resources etc. are super welcome!!

(I am starting to study via the book Make: Electronics, I hope to one day be able to help other people too, rather than asking so many questions!)

edit: typo
 

[Nick H]

5 day bump =)

Maybe if a mod sees and thinks this post would have been more appropriate in the beginners section, it might do better there??

Any pointers or tips are appreciated!

[tunk]

I suggest that you first open the battery and measure the voltage of each cell.

[tooki]

What makes you so sure it’s a 20S battery?

I googled a bit and found that often, they showed a picture of a 20S pack, but battery listings frequently use generic images. Other listings showed packs that definitely had only 15 cells, which is what 18V would be.

If it’s a 2000mAh pack, and it takes 10h to charge in the vacuum, then it’s using a 200mA charge current (i.e. C/10 rate).

To charge NiMH, you charge at constant current until the voltage peaks and then begins to fall*. For most NiMH cells, the final charge voltage is around 1.65V, though it can be as high as 1.9V. Assuming a 1.65V end of charge voltage, then you need around 25V.

24V would get you close, but the bigger issue is that it’s not constant-current, and a simple resistor won’t do that. A standard bench lab power supply will work, but only if you can actually watch it and eventually cut off the charge once the voltage begins to drop.

If you can’t monitor it, then you could do a very low current trickle charge (e.g. C/20, which would be 100mA here, or even less) since it can tolerate that.

Yes, a simple diode is all you need. Just bear in mind the diode voltage drop, so your maximum input voltage needs to reflect that. (Hence why a schottky is often used for this.)

*Technically, it’s a sharp increase in the rate of cell temperature rise that indicates it’s full. The increase in temperature causes an increase in the cell’s internal resistance, which in turn limits current slightly, resulting in a drop in voltage, allowing one to determine end of charge by the voltage change.

[Nick H]

Thank you very much for your reply!! I appreciate the help, I'd love to get the thing working.

What makes you so sure it’s a 20S battery?

I have the battery out of the hoover and there are definitely 20 cells, two rows of 10 - it's the same battery as the image I posted.

I googled a bit and found that often, they showed a picture of a 20S pack, but battery listings frequently use generic images. Other listings showed packs that definitely had only 15 cells, which is what 18V would be.

I also saw a couple of those 15 cell packs, 8 on the bottom with 7 on top, but they were all generic packs. The official replacement part is the one I posted, and the same as in the hoover with the 10 cells down each side. That is why I guessed that the board has the relevant step down converters on it, depending on how it's configured it gives out different voltages from the same pack for different models of hoover.

Didn't you also think that it was odd that the charger is 36v??

If it’s a 2000mAh pack, and it takes 10h to charge in the vacuum, then it’s using a 200mA charge current (i.e. C/10 rate).

To charge NiMH, you charge at constant current until the voltage peaks and then begins to fall*. For most NiMH cells, the final charge voltage is around 1.65V, though it can be as high as 1.9V. Assuming a 1.65V end of charge voltage, then you need around 25V.

So the charging rate and method I'd also understood as you explained here.

However all the information I'd found on NiMh until now has said that the max voltage was 1.4?? The cells in the hoover pack are sub-c size
https://eu.industrial.panasonic.com/sites/default/pidseu/files/downloads/files/ni-mh-handbook-2014_interactive.pdf

If you could show me where I could read a bit more about the voltages which you talked about that would be appreciated.

24V would get you close, but the bigger issue is that it’s not constant-current, and a simple resistor won’t do that. A standard bench lab power supply will work, but only if you can actually watch it and eventually cut off the charge once the voltage begins to drop.

I obviously don't understand well enough haha, I was hoping the resistor would cut the mustard. I have actually bought a DIY bench supply kit and have the transformer for it salvaged from a UPS, but I want to get further with the learning before I construct it. This project is partly why I bought it....was hoping I might get the battery charging solution before [hopefully successfully] completing the kit.

If you can’t monitor it, then you could do a very low current trickle charge (e.g. C/20, which would be 100mA here, or even less) since it can tolerate that.

Yes, a simple diode is all you need. Just bear in mind the diode voltage drop, so your maximum input voltage needs to reflect that. (Hence why a schottky is often used for this.)

So, sorry, just to be clear; can I or can I not use the 24v DC transformer I have with an appropriate diode?? I cannot, right, since it is 600ma?? And using that 24v DC transformer with a diode and a resistor is not enough to limit the current in a simple way to 100ma and allow me to do a charge under observation (ocasionally checking voltage, and regulargly making sure nothing is getting too hot haha)??

If I can't use the transformer I have, I wondered if you might have any suggestions?   

*Technically, it’s a sharp increase in the rate of cell temperature rise that indicates it’s full. The increase in temperature causes an increase in the cell’s internal resistance, which in turn limits current slightly, resulting in a drop in voltage, allowing one to determine end of charge by the voltage change.

I guess for me the main thing with charging the pack is just to get the hoover running once properly so that I can see everything else is working. If that is the case, I will check out the next step options which could include; replacing the pack, trying to upgrade it to Li-Ion (obviously a much bigger project as it would have to charge from a different circuit etc.), rebuilding another 20s sub-c NiMH pack with better capacity (there are much better cells on nkon.nl)....but the first step is just to charge this pack once.

Maybe I'll come across a higher voltage dc transformer at some point soon, easier now we're out of the strict confinement. I guess that would be much easier to step down the voltage....Or maybe I'll buy a boost converter. I was hoping to not spend any money As always haha

Thanks again for the help!!

[Nick H]

I suggest that you first open the battery and measure the voltage of each cell.

Hi! Thanks or the suggestion.

Would this suggestion be as a safety precaution before attempting to charge it?? Or what would your thought process be??

If it was possible to avoid doing so it'd be preferable since I don't have any of that nice shrink wrap to recover it with afterwards. It was for this reason I was thinking of trying to do a very low current charge and keep an eye on the pack, regularly monitoring the temperature of the cells to make sure nothing was getting hot.

Reviving this pack would be the easiest and cheapest overall solution, but my first hope is just to get enough charge into it (very slowly) to see the hoover working, or to see that it doesn't work correctly.

[tunk]

Mostly to see how (un)healthy the battery is - if some of cells
are 0 volt then you may not be able to revive it. Or if several cells
doesn't take charge the others may be overcharged. Maybe you could
use a needle to make small holes in the shrink wrap and then measure
the voltages. That said, I guess your approach is ok, but maybe first
use a lower current than 100mA to see how/if it takes charge. And if
you haven't already done it, calculate the required wattage of the resistor.

[tooki]

Thank you very much for your reply!! I appreciate the help, I'd love to get the thing working.

What makes you so sure it’s a 20S battery?

I have the battery out of the hoover and there are definitely 20 cells, two rows of 10 - it's the same battery as the image I posted.

I googled a bit and found that often, they showed a picture of a 20S pack, but battery listings frequently use generic images. Other listings showed packs that definitely had only 15 cells, which is what 18V would be.

I also saw a couple of those 15 cell packs, 8 on the bottom with 7 on top, but they were all generic packs. The official replacement part is the one I posted, and the same as in the hoover with the 10 cells down each side. That is why I guessed that the board has the relevant step down converters on it, depending on how it's configured it gives out different voltages from the same pack for different models of hoover.

Didn't you also think that it was odd that the charger is 36v??

If it’s a 2000mAh pack, and it takes 10h to charge in the vacuum, then it’s using a 200mA charge current (i.e. C/10 rate).

To charge NiMH, you charge at constant current until the voltage peaks and then begins to fall*. For most NiMH cells, the final charge voltage is around 1.65V, though it can be as high as 1.9V. Assuming a 1.65V end of charge voltage, then you need around 25V.

So the charging rate and method I'd also understood as you explained here.

However all the information I'd found on NiMh until now has said that the max voltage was 1.4?? The cells in the hoover pack are sub-c size
https://eu.industrial.panasonic.com/sites/default/pidseu/files/downloads/files/ni-mh-handbook-2014_interactive.pdf

If you could show me where I could read a bit more about the voltages which you talked about that would be appreciated.

24V would get you close, but the bigger issue is that it’s not constant-current, and a simple resistor won’t do that. A standard bench lab power supply will work, but only if you can actually watch it and eventually cut off the charge once the voltage begins to drop.

I obviously don't understand well enough haha, I was hoping the resistor would cut the mustard. I have actually bought a DIY bench supply kit and have the transformer for it salvaged from a UPS, but I want to get further with the learning before I construct it. This project is partly why I bought it....was hoping I might get the battery charging solution before [hopefully successfully] completing the kit.

If you can’t monitor it, then you could do a very low current trickle charge (e.g. C/20, which would be 100mA here, or even less) since it can tolerate that.

Yes, a simple diode is all you need. Just bear in mind the diode voltage drop, so your maximum input voltage needs to reflect that. (Hence why a schottky is often used for this.)

So, sorry, just to be clear; can I or can I not use the 24v DC transformer I have with an appropriate diode?? I cannot, right, since it is 600ma?? And using that 24v DC transformer with a diode and a resistor is not enough to limit the current in a simple way to 100ma and allow me to do a charge under observation (ocasionally checking voltage, and regulargly making sure nothing is getting too hot haha)??

If I can't use the transformer I have, I wondered if you might have any suggestions?   

*Technically, it’s a sharp increase in the rate of cell temperature rise that indicates it’s full. The increase in temperature causes an increase in the cell’s internal resistance, which in turn limits current slightly, resulting in a drop in voltage, allowing one to determine end of charge by the voltage change.

I guess for me the main thing with charging the pack is just to get the hoover running once properly so that I can see everything else is working. If that is the case, I will check out the next step options which could include; replacing the pack, trying to upgrade it to Li-Ion (obviously a much bigger project as it would have to charge from a different circuit etc.), rebuilding another 20s sub-c NiMH pack with better capacity (there are much better cells on nkon.nl)....but the first step is just to charge this pack once.

Maybe I'll come across a higher voltage dc transformer at some point soon, easier now we're out of the strict confinement. I guess that would be much easier to step down the voltage....Or maybe I'll buy a boost converter. I was hoping to not spend any money As always haha

Thanks again for the help!!

I’m replying on my phone, where correctly quoting section by section is a monumental pain in the ass, so I’ll just list my replies to your questions above.

- Very odd that they advertise it as 18V but use a 20S battery. Is there any chance at all that it’s actually a complex battery (like one 15S and one 5S in the same package)?

- In modern electronics, it’s not odd to see higher voltages, since we often use DC-DC converters inside, and using a higher voltage means you can carry the same amount of energy on thinner wire than using a lower voltage. (This is why USB Power Delivery exists, to allow devices to request a lot more power, while not needing thicker cable.)

- 1.4V is a typical maximum voltage of a charged NiMH cell. The maximum charging voltage is not talking about the same thing.

- Look at the charging characteristic curves in that PDF, which describes it all very well. Most of the cells exceed 1.6V when charging in the cold. And on page 19, which is where they explain charging, point 4 says to use a maximum voltage cutoff of 1.8V.

I learned a lot about NiMH charging from building up some charger circuits using the MAX713 integrated circuit. Its datasheet (which is IMHO a model of clarity, other than that it now refers to “linear mode” despite switch mode having been entirely removed; so when reading, when it says that, understand it as the only mode now available) goes into a lot of detail on the process and parameters.

A diode is all you need to protect the power supply, but it still has to be a power supply with a constant-current mode. A resistor alone cannot form a constant current supply. (You can use a resistor to limit current for an LED, for example, because the LED has a defined voltage drop. But the battery’s voltage drop would be changing throughout the charging process, so with a resistor, the charging current would thus change as well.)

Given that a bench power supply is one of the most essential bits of electronics gear, I’d get that up and running first. (Current limiting is a key feature, since one commonly sets a current limit to protect a circuit on initial power up after construction, so you don’t fry it if you made a mistake. Just one of countless situations where an adjustable current limit is needed.)

[Nick H]

OK gents, thank you both for the suggestions!

I wondered if it could be a complicated 15s/5s or something, too, but since there are only the two fat cables, and no strange bumps or irregularities in the pack I guessed not.

I love the needle hole in the pack suggestion - I already broke a bit to see if the cells were branded and discovered that the wrap holds up really well to small holes, but I hadn't thought of it to measure individual cells Easy to cover safely again afterwards.


I'm going to go back and have a proper look at the PDF - sorry to make you read it without having looked at it in more detail myself. I really had it in my head  so I wasn't seeing anything other than the 1.4, I'll go and study it properly. I'll also check out the datasheet you suggested

I reckon I'll  try and get the bench power supply constructed, it'll be a good learning process in itself and if it goes well I'll get a useful thing out of it I'll probably have to start a different thread to ask some questions at some point haha, the good thing is that the kit is very well documented [if not very favourably in terms of reviews of the kit, I bought it on a bit of a whim as it was 3 for 12e] with the common problems/solutions etc. it is the 30v 3a kit which has been around for a while. 

So for now I probably won't have any more questions on this thread for a while, until I've done some more studying and had the time to take the project a bit further.

Thanks for all the help so far it's really appreciated....in my circle I don't really have anyone I can ask face to face, and in the current situation that has been difficult.

[tooki]

I know exactly what you mean about having nobody to ask!

Anyway, don’t worry, you didn’t “make me” read the PDF. I just skimmed it, since I knew what I was looking for I was able to find it quickly. And then I saved it, since it looks like a great resource.