0
\$\begingroup\$

I've been working on a quadruped project, and need some help determining the... safety... of my power system.

To start, there's two packs, a larger "main" and a smaller "reserve"

the two packs are built from the same type of battery: Coolplay 3.7V 500mAh LiPo

the smaller pack is two of these wired in series to get 7.4V 1Ah, and the larger one is 8 of these wired to 7.4v 2Ah (4 packs in parallel to make 4x 3.7V 1Ah, in which they are series paired into 2x 7.4v 1Ah, which in turn are wired in parallel to get 7.4v 2Ah)

I've done the prerequisite searching, and can only find safe current to apply to parts and IC's, not the safe current to pull from the battery. So, my question is, what would be a "safe" amount of current to pull from either of these packs?

At idle (all IC's powered, but servos not powered), the build pulls about 0.2A, and the packs don't heat up at all, but when the build enters a "pose" such as standing (all servos powered up and under load), the load jumps up to 2A usage, and the pack temperatures start to rise, and my overcurrent kicks on and shuts the packs off (I have it set very low because, well I don't trust LiPos).

So, is it safe to run my pack under these conditions? and if so, for how long? I have adjustable overcurrent protection and temperature monitoring installed, so I can utilize those as well.

And if its not safe to run like this, what could I change?

One of these packs powers a Syma X5C drone for 7 minutes. So some quick and dirty math tells me that the drone pulls around 4.5 amps from the packs for the duration of the 7 minutes, and the packs are pretty warm from this. So with the new information that these individual batteries get 4.5A pulled from them, what is the verdict on this?

**EDIT: RESOLUTION So, thanks to @MarcusMUller and his insight, I've determined that the packs should be safe to run in these conditions but should be monitored more closely (as marcus pointed out, cheap batteries are unreliable and inconsistent) and should be housed better. Thank you @MarcusMuller for the feedback and insight

**NOTE: Batteries are dangerous, but I dont need to say that. As marcus pointed out (more or less), batteries with poor documentation should make you nervous. Good batteries are a good investment, and shortcuts should only be taken by those that know the path, and even then it should make you nervous. Be safe with batteries

\$\endgroup\$
12
  • 1
    \$\begingroup\$ The maximum permissible continuous draw is a pretty important property of a battery pack. Your supplier should be able to tell you; that's not up to you or us to know; overcurrent can both have immediately destructive (explosive, even) effect, as well as latent, capacity- and lifetime-decreasing effect (mostly: if the electrode is not designed for high current draw, you'd form metal spikes where there shouldn't be any). The first effects are easy to test for (increase current draw until the thing gets too hot, then dispose of it, use an identical one with at most x% of the last current). \$\endgroup\$
    – Marcus Müller
    Jan 3, 2020 at 21:10
  • \$\begingroup\$ the second needs cycle testing under lab conditions. It can be somewhat modeled knowing the properties of the electrode and the electrolyte and its carrier. That's really something only the manufacturer / their test labs knows. \$\endgroup\$
    – Marcus Müller
    Jan 3, 2020 at 21:14
  • 1
    \$\begingroup\$ I'm voting to close this question as off-topic because OP has bought a device from a supplier that doesn't supply a datasheet. Device doesn't work as expected for application. Since no guarantees made by a datasheet can be expected, there's nothing to answer here: Get a device that comes with a datasheet that guarantees operation under the conditions that you need. \$\endgroup\$
    – Marcus Müller
    Jan 3, 2020 at 21:14
  • 2
    \$\begingroup\$ How good is the ventilation? Does the battery have active cooling? "the smaller pack is two of these wired in series to get 7.4V 1Ah" does not compute. Do you mean 4 cells? \$\endgroup\$
    – Bruce Abbott
    Jan 3, 2020 at 21:15
  • \$\begingroup\$ the packs are cooled from the same fan the power regulator is cooled from. the packs i purchased mentioned no maximum draw but they are quadcopter batteries and i know those run up to 2A for around 10 minutes \$\endgroup\$
    – Werlious
    Jan 3, 2020 at 22:26

3 Answers 3

Reset to default
0
\$\begingroup\$

The Lipo batteries used in drones are usually rated for at least 10C, which means the maximium safe current is 10 times the Ah capacity. Your main pack has a capacity of 2Ah, so at 2A it is only being drained at 1C. Thus it would seem to be quite safe.

However if current is being drawn in pulses the rms current will be higher than the measured average current, which will heat the battery more. This shouldn't be a problem at 1C though, so long as the battery has adequate cooling.

Ultimately the most important factor affecting Lipo safety is temperature. Lipo cells are commonly rated for a maximum temperature of 45 °C when charging and 60 °C during discharge. At some temperature above this they will suffer thermal runaway and explode, so you don't want to go anywhere near there. Provided your batteries stay below 45 °C it should be safe, but let them cool down before recharging.

\$\endgroup\$
5
  • \$\begingroup\$ Thank you for the insight. I have temp sensors installed, but i was thinking of upgrading them. Currently its 3 thermisters inside the packs and one on the chassis for reference, Im thinking of getting an IC and including it in the new pack design \$\endgroup\$
    – Werlious
    Jan 3, 2020 at 23:41
  • \$\begingroup\$ If you are making a new pack it might be better to use a smaller number of higher capacity cells (eg. 2 x 2000mAh), or even just buy a 7.4V 'RC hobby' pack. These are usually rated for 25C or more, and hardly get warm at 2A. \$\endgroup\$
    – Bruce Abbott
    Jan 3, 2020 at 23:49
  • \$\begingroup\$ Im not sure if im going to get new batteries themselves, i was redesigning the pack their in. I chose a bunch of small .5Ah packs for this reason, and for the fact that they were small. The pack their in currently only has one end of the batteries exposed to airflow (the side the leads come out) and the other ends are pushed into a foam slot about 12mm. I want to make a new pack out of tin with a kind of "central" airflow source, similar to how HVAC systems are set up in buildings so the batteries get even share of cool air instead of the ones on the end getting the hot overdraft of the first \$\endgroup\$
    – Werlious
    Jan 3, 2020 at 23:57
  • \$\begingroup\$ Foam is a thermal insulator, which is bad. Use aluminium rather than 'tin' - it transfers heat better. Try to get air flowing around the battery from one end to the other. \$\endgroup\$
    – Bruce Abbott
    Jan 4, 2020 at 0:07
  • \$\begingroup\$ I have tin on hand so its easier for me to put some 1mm I already have on my bend press. Im gonna put them in a tin box in a 2x5 array and duct the fan airflow over the sides of each battery evenly, and PLA forks to hold them up \$\endgroup\$
    – Werlious
    Jan 4, 2020 at 0:27
1
\$\begingroup\$

Barring better information: If you know that someone is using safely using the battery packs for 7 minutes at 4.5 A, I think it's pretty safe to continuously draw 2 A – modelling the internal resistance as mostly ohmic, that's not even ¼ of the heat that gets generated at 4.5 A.

But you ask to verify the safety of your system, and that's impossible without a manufacturer datasheet guaranteeing that the batteries are operable under the conditions you want; no amount of "referene" usage is going to change that, as you don't know the lifetime / temperature / safety measures that other user (in your case, a consumer drone manufacturer) was assuming. And seriously, a USD38 drone isn't built to last.

In fact, the information that the same battery pack is used in a least-cost drone means that these packs are pretty certainly assembled from the cheapest cells available at the time of manufacture – which means that you can make absolutely no assumptions based on your battery pack having the same label as some other pack.

Still, 2 A isn't "much" for a drone-targeting battery pack. It "should" be easy to build reliable battery packs for a load like that. But: since it's easy, it should also be easy to give you hard facts on the admissible current draw, and since you can't seem to find such, it negates this assumption.

\$\endgroup\$
10
  • \$\begingroup\$ Thank you. This was incredibly useful insight. I think I'm going to stick with the cheap packs I have, and get some better temp sensors and maybe a dedicated fan for the battery pack. I'm also most likely gonna upgrade to isolated tin casings for the batteries (incase one spectacularly fails theres a lower chance of chain reaction) as currently there only about 3mm from eachother with the upper half in the fans cooling path and the other half sunk into the holder. Any thoughts on this? Im also accepting your answer and I'll update the question with my solution \$\endgroup\$
    – Werlious
    Jan 3, 2020 at 22:59
  • \$\begingroup\$ Sounds like investing in better batteries, which come with a datasheet, would have the higher return on investment, to be honest. \$\endgroup\$
    – Marcus Müller
    Jan 3, 2020 at 23:02
  • \$\begingroup\$ You have a good point. I'll probably end up doing it all, since I really dont trust LiPo I'll invest in a good NiMh pack \$\endgroup\$
    – Werlious
    Jan 3, 2020 at 23:09
  • \$\begingroup\$ I don't think it's worth downgrading to NiMh, to be honest. Compared to modern LiPo, NiMh sucks in most aspects, and your 2 A really aren't hard. \$\endgroup\$
    – Marcus Müller
    Jan 3, 2020 at 23:11
  • \$\begingroup\$ Seriously, you can build safe battery packs. Look at this boring samsung cell; you can buy it for ca 4€ all over the world (make sure to not get counterfeits), and since you already have thermal monitoring, you can really make a safe pack out of these if you just add overcurrent limiting: That thing doesn't like being discharged at more than 20A. You'll be fine with 2A. \$\endgroup\$
    – Marcus Müller
    Jan 3, 2020 at 23:26
0
\$\begingroup\$

Normally batteries are rated in terms of multiple of capacity discharge; a 5C 2.8Ah battery would be dischargeable at 14A. There are loads of 18650 cells that can achieve this. I've seen up to 30C advertised.

You can buy 18650 cells for like 2$ for a decent brand name one (Panasonic, Samsung etc), and you'll probably find the power to weight and power to volume is higher. 2x18650 cells will give you 7.4V at 2-3Ah in series, in 36mmx18mmx65mm = 42ml, your el cheapo packs, from what i can tell are 37x24x7mm, 10of so 62ml for the same capacity. Generally, bigger cells have better space: energy efficiency because they have less casing and wiring.

Temperature from over current or over charging is what will make them bang so add a thermal fuse for your rated current and the data sheet rated temperature, make sure any short can only happen with the fuse in circuit, fix the fuse well to the battery... Have a good think about ways it can short circuit or puncture... You'll be OK. They put these things in no end of consumer products.

You could also use a battery management chip to monitor temperature, current and provide a shut off.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.