Powered Up Lithium DIY

[ejayb]

Today I made a quick and easy adapter to run my Powered Up hub from a lithium battery. 

Information for this project came from Philo and Lygte.

 

The battery:

The Znter 9v LiIon battery is rated at 5.4wH. I choose this over the ZIPPY Lipo that Philo used because:

• It can be recharged via Micro USB
• It has a constant 9.5v output which means train motors should run at a constant speed regardless of charge level

It is available on AliExpress, I can post a link to the seller I used if allowed.

 

 

I cut a piece of Veroboard to fit the width and height of the AAA insert. Then I soldered a 9v battery clip to the columns that touch the pins inside the hub and scratched away the copper in the middle. It took me a couple of attempts to find the best holes for the clip wires. And I added extra solder at the bottom to push against the pins.

 

The battery and Veroboard fit tightly so no extra padding is needed to keep them in place.

 

The result:

I haven't had time to do much testing and I only have a single hub so I can't do a side by side. The hub powers up and works as expected. This could be a placebo but I feel like the train pulls better at low speeds and don't loose as much speed through the corners. This may be because lithium batteries handle high loads better than alkaline so the voltage don't drop when the motor is working harder. 

Edited February 22, 2021 by ejayb
Added second photo. Changed 9 to 9.5v

[LegoDW]

I an not that familiar with Lithium Batteries.  Why would you do this vs using rechargeable Ni-Mh?  I have been happy with my AAA Panasonic Eneloop rechargeable ones.  

[ejayb]

There are two main reasons LegoDW.

Cost:

I paid 12.48€ for two of the Znter batteries. The clips and Veroboard should could a couple of euros at most. 

12 750mAh Eneloop cells would have cost me 30€, plus I would need a AAA charger.

Voltage:

a) Rechargeable NiMh AAA cells have are rated at 1.2v. 6 * 1.2 = 7.2v. A Locomotive with fully charged Eneloops with run slower than with Alkaline or Lithium.

b) A NiMh and Alkaline battery (made of 6 cells in series) will rapidly drop from 7.2v and 9v respectively as you begin to use them. This results in your locomotive running slower and slower. The Znter battery will always output 9.5v.

In the comparison above, found here, The smaller 400mAh Znter can be seen delivering 9.4v for 300mAh. A 9v alkaline battery from Ikea starts at 9v, drops rapidly to 7.8 and is then at 6.7v by the time it gets to 300mAh.

(I realise this graph isn't a comparison of Eneloops/6*AAA and the Znter but it does represent the voltage drop well, regardless of capacity)

 

On the other hand,

Capacity:

The Znter is rated for 600mAh but I imagine testing would show 500mAh.

Eneloops are rated for 750mAh, so in theory the Eneloops should run a train for longer (although much slower for the majority of the time).

 

 

Edited February 22, 2021 by ejayb
Note regarding AAA vs Znter. Changed 9 to 9.5v

[Hod Carrier]

Are these batteries safe for novice use? In other words, do they have the necessary protections against mishap built-in?

[ejayb]

1 hour ago, Hod Carrier said:

Are these batteries safe for novice use? In other words, do they have the necessary protections against mishap built-in?

I believe so. They have a low voltage cutoff and an active charging circuit. 

Lygte-info carries out very detailed tests on many charging and battery related items and mentions no safety issues with them  

https://lygte-info.dk/info/BatteryDisassembly9VLiIon UK.html

[LegoDW]

Thank you for the explanation.  Was unaware the voltage on my Ni-MH was only 1.2 V.  Can see some advantages.  One key performance I see from the Power Up system, like the PF before, was putting the battery box over the motor.  How dose one 9 V battery compare to 6 AAA in weight?

[dtomsen]

Ansmann AAA/AA NiZn rechargeable batteries are norminal 1,6v and 550/1.500 mAh each.

Edited February 22, 2021 by dtomsen

[ejayb]

3 hours ago, LegoDW said:

Thank you for the explanation.  Was unaware the voltage on my Ni-MH was only 1.2 V.  Can see some advantages.  One key performance I see from the Power Up system, like the PF before, was putting the battery box over the motor.  How dose one 9 V battery compare to 6 AAA in weight?

The battery cradle with 6 AAA cells is 80 grams, the Znter battery with the Veroboard is 38 grams.

The Znter battery is smaller and you could ballast the cover or leave the cover off and lower the box by one brick.

 

Philo also has a graph showing the how voltage affects rpm.

[ejayb]

19 minutes ago, dtomsen said:

Ansmann AAA/AA NiZn rechargeable batteries are norminal 1,6v and 550/1.500 mAh each.

NiZn is probably the best drop in solution but expensive if you include the charger.

If I could borrow a set, I would do a comparison with the Znter. 

[dtomsen]

3 hours ago, ejayb said:

NiZn is probably the best drop in solution but expensive if you include the charger. 

Yeah, the charger is pretty expensive but not the batteries.

And beware the total voltage which is in the vicinity of 1,8-1,9v per battery or 10,8-11,4v in total the first couple of minutes after fully recharging.

The LEGO PF motors can handle this for the short duration but older components and other brands might struggle and need the batteries to discharge a bit 

Edited February 22, 2021 by dtomsen

[legotownlinz]

I don't consider the lower voltage of NiMh batteries a serious issue when using the train motor because the train motors are too fast. The higher capacity is more important. The only exception is the Crocodile with the slow Technic motor, I'm planning to replace the NiMh batteries there.

It has certainly be mentions several times in this forum, but I would like to repeat it again since it saves lots of money: The IKEA Ladda NiMh recharchable batteries come from the same factory as Eneloops and have an identical performance.

 

[ejayb]

14 hours ago, legotownlinz said:

The IKEA Ladda NiMh

Two sets of Ladda’s would still cost more than two Znter batteries and that’s before the cost of a charger (which is less convenient). 
 

14 hours ago, legotownlinz said:

the lower voltage of NiMh batteries a serious issue when using the train motor because the train motors are too fast

True, the train motor will easily derail at somewhere around power level 75 with the Znter. But for lower speed shunting, the consistency is nicer with a constant voltage. 

[UltraViolet]

  There are a number of brands offering similar batteries, and if I buy them I will probably be going through eBay.  (I even saw one type that can use an OTG adapter to reverse-charge another device out the USB, acting like a typical USB Power Bank.)  The most important feature to me would be the constant 9.x volt output level, followed by USB charging input - Do you think the Znter is the best overall when considering these together with capacity?

  Is there anyone else out there who has tested more brands of these batteries and found better alternatives?  Please comment also on similar AA-format batteries if you have experiences to share, as I am also interested in those.  My primary reason for looking into Li-ion compatible cells is to avoid the problem with leaking Alkalines.  I've had a TON of these AA's/AAA's leak randomly over the years, some within a few weeks of installation.  It's so random, messy and potentially damaging, I've had to resort to keeping all battery boxes and remotes empty between uses.  (The same goes for my wireless computer mice.)  I want the same voltage as alkaline while also having proper recharge capability, but with out the damned leaking!

[TuffTuffTuff]

Keybrick One gives you constant output using a LiPo battery. I hope it will be available again soon.

[Pendra37]

You can find a lot of different high capacity 14500 LiPo batteries. 14500 is AA form factor. Some even come with USB charging option.1 such battery can output 3.7V, which is equivalent to 3 rechargeable AA batteries. My only problem with LiPos is that they are not the most stable things around. They can catch fire if overcharged or overdrained. You need to be careful with them. The other option is 14500 LiFePo 3.3V. They are extremely stable and rugged but they only have ~500 mAh capacity. 1000 if you set them up parallel. 

[dr_spock]

Stumbled across an interesting video on the innard of the Znter 9V battery: 

 

[dtomsen]

To do something similar with PF, one can use two genuine LEGO parts, the old 9v battery box together with the 9v to PF extension wire

Voilà, a cheap and smaller rechargeable 1000 mAh battery box with high and stable voltage output  

Edited March 1, 2021 by dtomsen

[UltraViolet]

10 hours ago, dtomsen said:

To do something similar with PF, one can use two genuine LEGO parts, the old 9v battery box together with the 9v to PF extension wire

Voilà, a cheap and smaller rechargeable 1000 mAh battery box with high and stable voltage output  

  That is precisely what I had in mind!  (Modifying the cable or using other types of pre-fab conversion cables gives additional possibilities for destination also.)

Quote

Stumbled across an interesting video on the innard of the Znter 9V battery

  Thanks, dr_spock, for finding that video.  It was amazing!  The Znter batteries are indeed, as said by the presenter, very "logical" in construction.  I really appreciate anything that can disassemble/reassemble with screws and no permanent damage.  The in-depth description of the circuit and the components was quite refreshing to see, and gives me a lot more confidence in the battery brand.  One of the more intriguing details mentioned is that the internal circuit will slowly but eventually drain the lithium packs when not in use.  It might be possible to modify these by adding a cell disconnect switch as an improvement to the design.  The space is tight, but maybe enough.

  I think I have little to lose by ordering a few of the Znter batteries to play around with.

Edited March 2, 2021 by UltraViolet
Additional response

[UltraViolet]

  I noticed that the original version of the Znter 9V had the micro USB connector at the bottom, while the current crop for sale seams to all have them at the top.  Someone commented on one of the teardown articles, that the new version no longer has a screw together casing - just force together or clips maybe.  Can anyone confirm this?

  I also see that Znter makes AA and AAA type, as well as C and D format.

[1963maniac]

On 3/1/2021 at 12:15 PM, dtomsen said:

To do something similar with PF, one can use two genuine LEGO parts, the old 9v battery box together with the 9v to PF extension wire

Voilà, a cheap and smaller rechargeable 1000 mAh battery box with high and stable voltage output  

Nice idea, I thought of it too, but it didn't work for me. So, if it works for you please show us how you did it.

[dtomsen]

4 hours ago, 1963maniac said:

Nice idea, I thought of it too, but it didn't work for me. So, if it works for you please show us how you did it.

What didn't work if you don't mind me asking?

[UltraViolet]

3 hours ago, 1963maniac said:

Nice idea, I thought of it too, but it didn't work for me. So, if it works for you please show us how you did it.

  If directly feeding a PF motor or lights, it will work.  If you want to supply battery power input to the PF IR Receiver, if won't work without a modified cable.  I've been digging deep into LEGO wiring theory recently, with particular attention paid to standard polarity on various devices and wires, so I can speak to this issue fairly competently.  I will attempt to illustrate how you can go about making this work:

  The 4760/73955 9V battery box has a simple on/off switch, which simplifies keeping track of the correct polarity.  While under most circumstances polarity would not be particularly important to observe, in the PF system the fixed 9V battery supply lines to the IR Receiver or the Servo Motor must maintain proper polarity, while the separate control lines can reverse without issue.  The PF connectors are polarized by their physical keying, but some caution still must be observed if using the 9V stud system headers so as to avoid back-feeding power from a second source into the IR Receiver or PF Battery Box output, or attaching multiple stud-connector headers in reverse relative polarity on the conductive plates.

  First we must note the polarity on this 9V battery box:

  With the switch end facing you, the left row of studs is positive.  When attaching the PF extension cable to this box using the 9V/PF connector end of the cable, the wire lead can face straight away from you and the switch, or towards the right side.  (Left or towards you makes the polarity backwards.)  I keep track on my 9V battery boxes simply by attaching a 1x1 red tile somewhere on the correct side of the box.

  With that established, we can now move on to modifying the cable:

  To observe correct polarity, the outer two conductors on each side can be stripped and soldered together somewhere along their length.  (Offset your stripped points slightly so you don't have to insulate the two soldered pairs from each other.)  GND and C1 will be one joined pair, while C2 and 9V will form the other.  (This layout was clearly created so that the outer pairs of wires don't cause damage if they somehow short together.)  Alternatively you could solder bridges on top or bottom of the PF end connector, or on the PF top part of the 9V/PF end connector.  Please note - I have seen an image posted joining C1 to 9V and GND to C2 - don't do this unless you absolutely remember to reverse the orientation of the 9V connector on top of the 9V battery box!  It is best to stick with the proper polarity I described in order to avoid potential damage and to not get confused later.

  I also strongly recommend not using either of the two later types of 9V studded battery boxes with direction switches/buttons, as this significantly adds to the confusion and risk if you don't properly keep track of what you are doing.  If you do so, be certain to label the positive studs in relation to the directional control.  If you want to be even more safe, or are simply paranoid, you could splice a protection diode of sufficient amperage handling in-line with the cable at the point where you pair the neighboring wires on the positive side pair.  As long as the polarity is correct, you can use the modified cable to power the IR Receiver or any PF motor/servo directly, with the rotation being in the expected direction (forward polarity equals clockwise rotation when the motor is viewed from the axle face).  You can then reverse a motor/servo using a PF polarity switch if not using an IR Receiver.  The PF polarity switch would have no effect on an IR Receiver if it were connected ahead of one, so it is not harmful.  If feeding a PF servo directly from the modified cable only, obviously this will result in full travel to one side or the other only because the control lines are fed full voltage without any PWM.

  One other method, as a bonus, is available to you if you own a PFx Brick.  (This may apply similarly to other third-party controllers, but I don't have any others yet to prove it.)  The PFx Brick has a cut-away in the bottom that will fit a PF connector flush.  The bottom connector is for 9V power input to run the Brick.  The designer smartly set up the input pins in such a way that power can come from the 9V/GND battery box contacts OR the C1/C2 control line contacts.  (C1/C2 must still maintain the standard forward polarity as detailed above.)  The advantage of this is that the ordinary 9V-to-PF cable can power the PFx Brick from a studded 9V battery box without any cable modifications!  An even neater trick I learned from this is that you can actually use just the 9V end of the cable sandwiched between the PFx Brick and the top of the 4760/73955 battery box.  The free end of the cable could quite simply be cut off.  That loose end of the cable can still be very useful, however, as a constant 9V supply for always-on lighting, or even to power a second PFx Brick!  See below a photo taken just now at my desk showing the stacked configuration and the green power LED lit on the PFx Brick:

  You can also see where I put the red tile on top of the battery box to indicate the always-positive side, and a blue one on the always-negative side for good measure!  (You can never be too sure. )  One thing that occurred to me about this stacked assembly is that you could use the pin holes in the side of the PFx Brick to support everything, meaning the battery box could be freely suspended.  The importance of this is that the battery cover will not be fouled on the bottom of a build like a train, so you can open the cover and swap out the 9V battery without dismantling anything else.  The studded clutch between the PFx Brick and the top of the battery box is extremely and sufficiently firm to maintain the suspension.  Use of lighter weight batteries, as is the case with most rechargables like the li-ion version under direct discussion in this thread, would only make this more assured.

Happy modding!

[TuffTuffTuff]

@UltraViolet great explanation and ideas, thank you

[dtomsen]

But wouldn't the 9v to PF wire solve the polarity issues?

[1963maniac]

On 3/3/2021 at 1:11 AM, UltraViolet said:

 If you want to supply battery power input to the PF IR Receiver, if won't work without a modified cable. 

Thanks for the great explanation. Much appreciated!! So, would you please explain the modified PF cable too. (some of us aren't such good electricians)

Edited March 5, 2021 by 1963maniac