CPAP gets 18650ed.

There is no actual communication except for overheating of the supply. It does sense for a 2.7K resistor being present to identify it is an 85W supply. Please read my previous post. This is exactly what you need to do.
 
I appreciate the information Norknid, but the adapter I have already does the 3.3V signal thing. It's made specifically for my cpap model. The main reason I bought it was for the compatible plug for the machine. I looked at all the parts to make a DIY unit, and buying this adapter/regulator ended up being cheaper.

I used my temporary battery and took a 2 hour nap this afternoon, with the cpap powered through the adapter. Works perfectly! I also made a diagram of all the stuff I'd like to fit in the box. No idea if it will all fit, but I'm thinking it would be great if everything I need is in the box and I just have to open it up and turn it on. Well see how that goes.


image_rzfidq.jpg
 
I assume you are talking about the battery converter box they sell for the CPAP. I assumed you were going direct from that 24V converter into the machine. For others who might read this thread, here is a solution that would have less to carry around. Cut the Resmed cable at a convenient location and install a three pin plug that allows you to put it back together. Then make another plug that will connect to your buck or boost converter. At the connector install the two resistor divider (prior post) that will simulate a 3.3V supply with a 2.7K ohm resistor in series to simulate a 85W AC power pack. NEVER connect the center pin directly to 3.3V as it will damage the machine. I never had any need to do this because these power packs are $10 and keep a couple spare. Mine is wired permanently to my camp with the converter 50 feet away.
 
Norknid,

I have the Resmed S9 unit, with a 3pin angled plug.
Do you know if this is wired the same or where to get the info?

Thanks
 
I think the convention is the same, just the plugs won't fit each other machine. I have no direct experience. If you connect a 2.7K resistor from center pin to common outside shell, it should measure about 1.6V. I wouldn't do this test without an insulated center connector. It would be too easy to short the 24V to the center pin possibly damaging the pack.
 
My BMS finally arrived from China! Still plugging away at testing cells though. So far, I've done around 70 or so, so I'm almost done. The best 66 will go in the battery.
I've realized I can't get a 4A fuse, so I'll make due with a 5A. Using the axial glass 0.5A fuses on each cell, that would allow the battery to deliver almost 11A before they start to blow. So a 5A main fuse will definitely pop before that. (I hope.)

I've been playing with an actual wiring diagram, to better imagine where all the parts will go. Here's what I've got so far:
https://i.pinimg.com/originals/75/be/49/75be49c37ec7f697bf060706130b692c.jpg

I definitely wanted a Voltmeter to show the battery status, but I decided to spend a couple extra bucks (I know, unusual for me.) on a Wattmeter. This will show Volts, Amps, Watts, and Whrs, to give me a better idea as to how the battery actually performs vs my math estimates. Plus, it has a low Voltage alarm that can be set. The BMS should shut it down automatically if the Voltage gets too low, but I've never really trusted BMSs.

If I can fit them in the box, I'd like to have the option of taking a battery charge controller and the AC adapter, in case I come across electrical outlets in my travels. Once I get the box, I'll be able to do a mockup of all the parts and see what will fit, and what physical shape I'll make the battery.
 
Last edited:
I've just realized that I used up all my copper wire on other projects! I'll have to get more before I can start building the battery.

Some good news though, I've finally finished testing all 99 cells, and they're all good! There's something very satisfying about seeing all the cells lined up together.

image_uyclvz.jpg


The range of mAh is pretty good. Nearly all are 2500mAh or higher. A few lower ones, but when I harvest random old laptop batteries, they're usually not this good.

image_zsqqyx.jpg


Anyway, I've chosen the best 66 to be in the CPAP battery. They'll range from 2661-2545mAh. That should give me some nice Watt-hours.

image_xqzped.jpg


I think it's about time to go get that box and see how much I can fit.
 
Excited following this thread. I haven't purchased a mobile source for my S9 Autoset so when camping have been bringing a 100' extension cord.

Is your machine also a bilevel machine and/or are your calculations going to be different than for my usage?

thank you
leo
 
Hi Leo,
Each machine will have it's own power requirements, so the values will be different. But the basic math could be applied to any device. The really important values are Volts and Amps that the machine needs, and Volts, Amps, and Amp-hours that each cell can safely deliver. I have the ResMed A10, which is a regular CPAP (or APAP, I don't know the difference, and I've heard it called both.)

I finally went and purchased the box that all this is going in. I don't have pictures yet, but it looks like everything I want in there is going to fit! Some bad news though, is that I crashed my computer last week, and I've been working all weekend to get it back up and running. Some good news, is that I had no data loss thanks to luck and backups. :)
 
Got my computer 95% up and running.

Now that all the cells are tested, I got them cataloged and entered into repacker. All arranged how they will be soldered up.

image_aimauc.jpg


I ran out of the wires I normally use, so I had to scavenge some from a box of junk wires I've got laying around. Took one multi-stranded wire apart to use individual strands to connect the bottom half of the battery to each cell. Another multi-stranded wire was the right size for the bus, but insulated. I just cut a little insulation off at each solder site, and it seems to work fine.

image_nltqro.jpg


BMS is installed. I guess it's working. At least it's not smoking, and the power out Voltage is right.

image_gziihf.jpg


This particular BMS has the option to add a thermocoupleswitch. So I added one that is normally closed until 113F. Still need to add a cable for balancing and a fuse, then some outer physical protection. But it's shaping up!


image_jvsbjj.jpg
 
I think having the BMS sitting on multiple cells without any padding/isolation is a potential hazard. This is going on road trips / camping trips and get shaken/bumped a lot along the way, right? Pointy parts on the back could rub through the cell wrappers and eventually cause a short.
 
It's there, just not pictured. :)

Also, I actually made a little progress with the box.


image_utnlkx.jpg


Unfortunately, the charging hardware doesn't fit. But all the stuff necessary for camping does. I still need to fit the power meter and switch in such a way that it doesn't flop around during travel. The battery fits pretty snug, but I'd also like to make sure it doesn't have space to move around after the box is shut.
 
Finished!

I finally took an afternoon to build the little power box.

image_ybergt.jpg


image_sfqbpw.jpg


image_tdleda.jpg


image_psgelx.jpg


The Watt meter will keep track of the total energy used, and there is a low Voltage alarm that I can set myself that should go off before the BMS low Voltage cutoff does. I think I'll keep the little battery checker in the box too. It gives some peace of mind knowing the series are all balanced.

I'm hoping to get around 32 hours of sleep between charges. I'll start using it tonight and see what happens. Also, someone asked how much the whole thing weighs. It is 21 lbs.
 
I have some early results from my first night on the CPAP battery. It looks like I may have overestimated the power requirements with the heat and humidifier off. According to the power meter, I used 42 Wh, and the CPAP itself counted up 7 hours 49 minutes. When I woke up, the meter was reading 0.3A instead of the 0.8A I was expecting.

So 42 Wh / 7.82 h = 5.37 Wh/h, approximately.
The battery should have about 632 Wh, so 632 Wh / 5.37 Wh/h = 117.6 hours of use.
That's 14.7 nights if I sleep 8 hours per night. Around 2 weeks! I was hoping for 4 days.

I'll keep using the battery until it's discharged, and see what the numbers are then. Hopefully my math's not wrong!

Also, for anyone who's curious, I ended up building this whole thing for about $180. I don't normally keep track of the cost of my projects. Takes some of the fun out of them. But in this case, I wanted to compare what I could do with what is available store-bought. The cheapest battery I could find would have cost around $250 per 8 hour night. And there wouldn't be a nice foam padded, water proof box either.
 
Your're right, with just the fan, the CPAP hardly makes a dent in the capacity! I tried running the humidifier/heated tubing too (in airplane mode). I can get 3 nights with the additional load, and the pack is in the 5% SOC range. My pack is a similar Wh capacity as your's, albeit a different voltage (663Wh, 7s8p).

So...
Fan only: 5.4W draw
Bells and whistles: 27.6W draw
 
I ran the CPAP off the battery for 12 days, I think. It shut off some time at night, but I was able to get some numbers.

It looks like one of my series discharged faster than the others. When it reached 2.5V the BMS did its job and shut the whole thing down. So I didn't get the full 632 Wh I was expecting. And there appears to be a fault somewhere in the battery.

Here's what I got though. The CPAP itself recorded 69.97 hours of total use. The energy meter recorded 412 Wh total. An average of 6.34W. I also noticed that the regulator uses about 1W even when the CPAP is unplugged. So that seems consistent with Lucan's observations.

My first thought about the battery is that one or more fuses might be blown or broken. That would reduce the capacity of a series, and those glass axial fuses are pretty fragile. After visually inspecting each one, and checking continuity between one cell and all the others, It seems the fuses are all intact. I'll have to open the other side of the battery and check the solder joints.

If I can't find anything physically wrong, I'll test each series individually for capacity. Depending on what I find, the whole battery may have to come apart. All these cells are the same make/model and all tested with similar capacities. I packed them with repacker, so it seems unlikely that the cells would be at fault. More likely I did something wrong. But I can't have a battery that's going imbalanced after a single discharge. I'll just have to test everything until I find the culprit.

But on the bright side, I got 12 nights out of this battery when I was initially expecting only 4. That means if I need to rebuild the battery, I can make it smaller and still have plenty of capacity.
 
On those LG cells the cutoff is 2.75 volts. When you read 2.5 volts, did it bounce back up with no load. I aim for 3 volt cutoff on my packs.

I would only take the pack apart if one of the cells is severely out of balance with the other cells and multiple charging discharging doesn't get it to balance. If you can't get a full charge even with a balance charger, then you might have a cell that is self discharging. All it takes is one bad cell to prevent the pack from being in balance.

Instead of the cellmeter8 I recommend the tenergy lipo checker cellmeter-6 (about 12 dollars on amazon), it's smaller and will give a loud alarm if the cell voltage gets too low (programmable) of if the cell voltage goes above 4.22 volts.If you do have a bad cell, you want something to alarm if the pack voltage goes too low. I had a pack that I didnt use for a few weeks, when I decided to use it , found it had drained completely, the cells were at 1 volt. If the pack had an alarm I could have saved the pack.


image_zattho.jpg
 
Yes, the low series eventually recovered to about 3V. If they had all discharged evenly, I would have heard an alarm at 16.8V for the batterty, or 2.8V per cell. I haven't determined yet why the one series discharged so much quicker than the rest. The battery was balanced when I started. The cell checker is just to give me a quick view of the individual series. I don't have it plugged in all the time. Fortunately, the BMS did its job and cut the battery off when it sensed one of the series was too low. Better 2.5V than 1V eh? :)
 
I got all 6 series tested on the CPAP battery, and here's the results:
1: 23.2 Ah
2: 27.1 Ah
3: 26.1 Ah
4: 15.2 Ah
5: 26.6 Ah
6: 18.5 Ah

Series 4 was definitely the one that caused the BMS to shut off the whole battery. But series 6 wouldn't have been far behind. I'm not sure why there is such a large variation in capacities. I'm definitely going to pull the battery apart and retest each cell individually.

When I rebuild it though, I'll be making it smaller. Probably 5p instead of 11p. That'll still give me about 5 nights sleep, reduce weight, and take up less space in the box.
 
You must log in or register to reply here.
Back
Top