diy solar

diy solar

Hot swapping LI-ON Tools battery circuit? Anyone using Blocking Diodes?

Chris Nafis

New Member
Joined
Feb 11, 2020
Messages
50
Location
Vischer Ferry NY
I've been using RYOBI 18/40V tool batteries to make powerstations. When putting batteries in parallel I currently have to make sure I'm using similar batteries (Ah) with equal charges. I don't want a huge rush of current from a fully charged battery to an empty battery. I noticed they make high voltage/high current Blocking Diodes. Has anyone used these? One concern I have is what will be the available current for the load. If I have two batteries. One full, One half-empty. Each has the blocking diode between the battery + and the load. If each battery can provide 10A of power, what will the available current be for the load? I ordered some diodes for an experiment... but thought someone might already know the answer. Thanks!
 
Those Ryobi Lion battery pack has BMS to protect the cells from overcharge/discharging, my 18V pack has it, never seen one without BMS.
 
Those Ryobi Lion battery pack has BMS to protect the cells from overcharge/discharging, my 18V pack has it, never seen one without BMS.
Yes the batteries have a BMS. Just didn't think they were designed for being placed in parallel. My RYOBI table saw takes two 18V batteries. I'm not sure if they have an additional circuit to handle mis-matched batteries...
 
Can you describe the circuit you have? What do you mean you're using them "to make powerstations"? I'm asking because often the tools which use two batteries (ex: 36V blower, miter saw, weed whacker, etc.) have the batteries wired in series, so there would be no such inrush current of having them in parallel. If you're talking about a particular Ryobi tool, you could try to find the electrical schematic / wiring diagram.
 
Last edited by a moderator:
I think he is building it based on this Ryobi power station that allow you to use your 40V power tool battery pack to run the inverter.
 
I see, so a battery bank with charger(s) and an inverter. So if you're making this out of removable Ryobi batteries (I would not recommend it), then yes I think there should be blocking diodes between paralleled batteries. But, that's probably not the right approach if you're going to make the batteries removable as then you're guaranteed that only the highest voltage battery can be drawn down.

I think @Bud Martin is correct that all tool batteries have their own BMS integrated inside, as this keeps the tool itself very simple (just a switch, motor+driver, over-temperature sensor and the cutting/functional end of the tool).

If you really want to DIY this, you should put a chip on each battery port which can deal with the voltages: sort of like a separate-input paralleled-output switching regulator. Give me 15 minutes and I'll try to find one online.
 
Actually I see those are 40V Ryobi batteries, so you're looking at 75-100V MOSFETs and this sort of thing would be a pcb-level design, not a chip/module you can just buy. I'll keep looking for a little longer for some off the shelf multiple-battery regulators like I described, but not sure there are many people with this same issue; most people on here aren't hot-swapping their batteries like tool batteries.

I'd like to reiterate that a DIY multiple-battery 40V, 6Ah system doesn't seem super safe if you are asking how diodes work. But yes, a ~10A 100V diode is totally reasonable and can be purchased for a couple dollars, it will just waste a lot of energy on heat compared to a better circuit that uses transistors with their gate boosted up. This generally requires an integrated circuit to control the FET.
 
Can you describe the circuit you have? What do you mean you're using them "to make powerstations"? I'm asking because often the tools which use two batteries (ex: 36V blower, miter saw, weed whacker, etc.) have the batteries wired in series, so there would be no such inrush current of having them in parallel. If you're talking about a particular Ryobi tool, you could try to find the electrical schematic / wiring diagram.
Ah that's right they are probably just using them is serial config.
 
Neat stuff, Chris!

If you want to fix the parallel issue, I think you should add some circuit at the point where you combine the 4 sections in parallel, for exactly the reason you noted in your first post. Your design is roughly 36V (let's say <= 48V; I don't know the Ryobi voltage details), and what is the current? If you can get 2600W from it, that must be roughly 650W per paralleled 2S pack, which is 18A per battery, or 20A if the inverter is 90% efficient. I assume these batteries can handle that current...you should check.

So you want a 4:1 combiner capable of 20A per leg and at least 48V. Actually now that I think about it 4x MOSFETs with an IC for the gate drive is not really what you want, as it would also result in the "back powering" issue where the most charged battery shoots a bunch of current back into the less-charged batteries when they get connected (even if connected "slowly" through FETs).

Diodes also would not work as I mentioned above because only the highest voltage battery pack would be providing the majority of the current, which means your 2600W system would only be a 650W system unless all 4 parallel legs were close to the same voltage.

So really what you want is a 4-input 1-output switching regulator, I think...need to think about this a bit more.
 
Neat stuff, Chris!

If you want to fix the parallel issue, I think you should add some circuit at the point where you combine the 4 sections in parallel, for exactly the reason you noted in your first post. Your design is roughly 36V (let's say <= 48V; I don't know the Ryobi voltage details), and what is the current? If you can get 2600W from it, that must be roughly 650W per paralleled 2S pack, which is 18A per battery, or 20A if the inverter is 90% efficient. I assume these batteries can handle that current...you should check.

So you want a 4:1 combiner capable of 20A per leg and at least 48V. Actually now that I think about it 4x MOSFETs with an IC for the gate drive is not really what you want, as it would also result in the "back powering" issue where the most charged battery shoots a bunch of current back into the less-charged batteries when they get connected (even if connected "slowly" through FETs).

Diodes also would not work as I mentioned above because only the highest voltage battery pack would be providing the majority of the current, which means your 2600W system would only be a 650W system unless all 4 parallel legs were close to the same voltage.

So really what you want is a 4-input 1-output switching regulator, I think...need to think about this a bit more.
Thanks! RYOBI came out with that new 1800W Powerstation. I think they fell short. Especially because they don't have pass-thru charging. I just determined that the 40V battery is capable of pass-thru charging... I made a single battery Powerstation, but I'd like to combine multiple batteries.
 
Ok, this is by no means a recommendation that you proceed with a real design (especially if you don't know what you're doing), but perhaps you could use something like a non-synchronous boost-mode SEPIC regulator on each of the paralleled battery paths. For example: several of these LT8710 devices with paralleled outputs (see Figure 16):


That non-synchronous boost mode SEPIC topology has a Schottky diode which would prevent the shared output node from back-powering any of the individual batteries, I think. And this particular ADI part has current-sense so you could ensure that each battery is not over-drawn (for example if you continue to draw a high load current when there are only a few batteries inserted). As an example you could boost up each 18V battery to whatever voltage you need for your inverter, or you can boost up your existing 36V 2S strings to something like 48V for the shared node. Your inverter would then run off that higher voltage. Inverters often come in 12V, 24V, or 48V flavors...are you using a specific 36V one to connect directly to the batteries?

A nice feature of using 18V for each boost regulator input is that you can then swap batteries one at a time and you don't have the two in series so they don't need to have approximately equal state of charge when inserted. Nor does their state of charge or voltage need to match that of any other paralleled section. But, the current from the batteries will be higher if it's a lower voltage going into the boost regulator and/or into the inverter. This can just be solved with more batteries and thicker wire though.


I had a second option listed here but I deleted it, as it was not a good idea.
 
Last edited by a moderator:
Synchronous mode (Figure 24, or the entirety of page 40) may work as well. You'd have to look at this more carefully as most datasheets don't show "paralleled boost converters with different input voltages" as a typical application. But, it can be done I think, it would just take some time to design and test it.
 
Back
Top