Switching from Nicad battery to Lipo: Am I doing this right?

[Nohippychicks]

BLUF:  I don't know how to troubleshoot a generic setup of Lipo cell ->charge controller-> boost converter->12v load and I'd like some help.

Long version:

I recently got 'hold of an old Yaesu UHF transceiver.  The battery is labeled as 12V, 500mAH and hasn't been charged in probably 30 years.

Thinking this would be a pretty fun little project to get working again, I tried to do a little homework online but didn't come up with exactly what I needed, so I went off past experience of a few other project where I simply powered a Raspberry Pi with a lithium cell and a charge controller; I figured I had all the know-how I needed and ordered up some parts. 

I got a (All generic) 3.7V, 4K mAH cell and a 5 pack of cheap TP4056 based charge controllers, as well as found a boost converter that I had left over from another project.

Everything arrived and I set about pulling out the old nicad cells and fitting the new bits into the battery housing.  After charging up the battery and loosely fitting everything in place, I adjusted the boost converter to 12.8v and the radio came to life.  Project done!  Only not so much.

Seemingly randomly, the radio would shut down.  A voltage check across the battery terminals shows 1.1v.  Following back, I get 1.1v at the output of the boost converter, 1.1 at the input of the boost converter, 1.1V at the output of the charge circuit and 4.x volts at the input of the charge circuit...4.xx volts at the output of the Lipo. 

This 1.1V output is intermittent enough that I can't figure out what the problem is.  Some times, I can remove and replace lipo cell and it's fixed for a while but some times I can measure 12v at the battery terminals but when I turn on the radio, I get a flash of life then nothing...and I'm back to measuring 1.1v at the terminals. 

Sometimes I can get the radio to turn on, where I just leave it for a few hours and everything is great, 'till I turn it off, then back on and back to 1.1v.  Sometimes turning off and on works as it should. Sometimes, if I hook the charge circuit up to USB, as if I were charging it, it pops back up to 12V and may stay there for a while or it may not.

Thinking it was the charge circuit because I got the cheapest ones I could, I swapped it with a new one from the same order (The only ones I have)...same deal...and a 3rd one, all with the same result. 

Since I get the 1.1v at the output of the charge circuits, I'm pretty sure that's the problem, but where is the solution?

Do I just buy a more expensive, and presumably better quality, one from a place like Adafruit?  Is there more troubleshooting that needs to be done?  As cheap as the chargers I've bought are, I'd hate to think that 3 (so far) out of 5 could be bad out of the box.  Maybe there's something more obvious that I'm missing.  I can post pictures of it all, but with the intermittent 12V I'm reasonably certain that everything is soldered up properly. 

[TERRA Operative]

How much current is the radio pulling from the battery?
Maybe the overrcurrent protection is kicking in in either the boost converter or the battery BMS.

Is the boost converter getting hot? They tend to use fake silicon, so replacing the boost chip with a genuine item and sticking a small heatsink on can go a long way.

[borghese]

What kind of battery do you have? If these are Chinese, they are likely to have only fractions of nominal capacity. A better way would be to use high capacity NiCd batteries; these are much easier to recharge, better tolerate charging abuses.

[Nohippychicks]

How much current is the radio pulling from the battery?
Maybe the overrcurrent protection is kicking ...

Is the boost converter getting hot? They tend to use fake silicon...

I just took a reading of .065 amps when I could finally get the radio to turn on for a few seconds.  I tried to hit the transmit button to see what it would pull then, but it shut down again and I haven't gotten it back up and going, but I'll update if I'm able to.

Does that fake silicone also go for the little charge controllers?  Maybe it's worth it to just buy a decent quality one instead of fooling with the cheapest ones I could have gotten...which is what I suspect.  Even more so after this.

The boost converter isn't getting hot at any point that I've found, but when I FIRST charged the new battery, the charge controller got uncomfortably hot.  I figured that it was because it was rated to charge at up to 1A and with a partly charged battery/cell.  I put a little heat sink on the chip, but eventually swapped that unit for a new one...then another new one.

I couldnt find a datasheet for the exact charge controller or battery protection that I have because of how generic I went, but at such a little draw, it doesn't seem like it should be tripping any sort of protection...but again, "Generic," so maybe?

What kind of battery do you have? If these are Chinese...

I do have a generic battery, but even if it's half of the rated capacity, it's still 4X what was there (500mAH).  Do you still think it's better to go back in with Nicads?  I don't have what it takes to string together my own pack of cells. I've found a service that would rebuild them for about $75, which is not really worth it to me, so I was trying to stay cheap.  Even if I go with a name brand lithium cell to try to avoid poorly labeled products, it seems like it'd be better, but I'm not sure.  If the consensus here is "Go back Nicad," I'll keep looking into them and get it done "Some day."

Thanks for the suggestions!  I'll wait to hear back about the lipo/nicad and go ahead and order some non-generic parts tomorrow.

[thm_w]

The capacity of the lithium cell is probably 1500-2000mAh at 3.8V, so ~8Wh. If you originally had 10 NiCd cells that were 500mAh, at 12V, that would be 6Wh. So the difference is not so large. But you could always parallel a few 18650's if you want greater capacity.

Is the circuit something like this:
https://www.aliexpress.com/item/5PCS-Micro-USB-5V-1A-18650-Lithium-Battery-Charger-Board-With-Protection-Module/32647794396.html

DW01A: http://cxem.net/master/files/97_DW01A-DS-11_EN.pdf with overcurrent 150mV
If the FET is 8205A: https://www.maritex.com.pl/product/attachment/91261/8205A.pdf
rds on about 25mOhm*2
So overcurrent trip point is roughly 3A. There is a chance the boost converter is pulling a lot for a short time, meaning the charge board shuts off (could be why you see 1V).
Its normal for the TP4056 to get hot as it is linear charge, so its dropping 5V - Vbat across.

Do you have a bench PSU you can test at 12V to check if that voltage is high enough, and what peak draw is?

[TERRA Operative]

The TP4056 lipo charger modules are pretty good, they work well enough.

It's the LM**** based buck and boost converter modules that are often using fake silicon. They work relatively ok, but replacing the chip with a genuine item makes them work faaar better and when you replace the caps with low ESR genuine types, they have a much nicer output.

Can you run your radio on a proper power supply and measure the current draw on transmit? That will tell you what your supply requirements are.

[Nohippychicks]

I do not have a bench supply yet.

The only way for me to run off of a proper supply will take some cobbling together back at the house.  My version of a "Proper," supply is a12V, 1A wall wart.  I will come up with something that takes fewer hands so I can power the radio AND transmit to pull a reading from it.  I'll also pull a reading for transmit and backlight on, just to see what the absolute most it may draw.

I'm almost certain that my boost converter has the fake chip if it's that common;  I bought these to play around with and went by price (Then by shipping time) on E-bay. 

Do you have a link to an article about those fake chips?  Doing a quick search here brings up some criticism of the soldering on them, but I think it'd be interesting to read a little more about that. 


I appreciate the help and the willingness to deal with me not having all the stuff I really should have to figure this stuff out.

[thm_w]

Is the 12V 1A regulated output? ie does it measure as 12V when plugged in or is it significantly higher.

If the 1A supply can easily handle the radio, seems less likely its overcurrent but still possible.
For example if the 18650 is pulling 3A at 3.8V (trip point), then you'd have at most about 10W or 12V at 0.8A on the boost output.

[KL27x]

Seemingly randomly, the radio would shut down.  A voltage check across the battery terminals shows 1.1v.  Following back, I get 1.1v at the output of the boost converter, 1.1 at the input of the boost converter, 1.1V at the output of the charge circuit and 4.x volts at the input of the charge circuit...4.xx volts at the output of the Lipo. 
Since I get the 1.1v at the output of the charge circuits

You have typoed something, perhaps?

4.xx volts at the output of the Lipo.

 

1.1 at the input of the boost converter,

...and

Since I get the 1.1v at the output of the charge circuits

AFAIC, the input of the boost converter and the output of the charge circuit and the voltage across the LiPo terminals is the same node. It can't be 4.xxV and 1.1V at the same time, unless you have some circuitry in there that is not implicit/expected.

That said, when you try to transmit, that takes power. If it was your boost converter that was the bottleneck, the inductor would saturate and the voltage would plummet. But you should probably measure at least 3.6-4.2V. If it were the battery's internal protection circuit, 1.1V sounds feasible as a leak or something. But I would be a bit surprised to see a 4Ah LiPo cell with max discharge rate less than 4x (to account for the boost) w/e current that old 12V 500mAh NiMh had supplied. 

So it's a weird thing you got there.


Edit: oh wait, it is NiCd. Those can discharge insane current.

Have you tested the radio on a lab PSU or whatnot? This where it is helpful to remove variables. Maybe the radio is damaged.

[Nohippychicks]

The battery I'm currently using is lithium; the original battery arrangement that I'm looking to replace was nicad.

The setup that I am trying goes: Lithium cell to a charge controller to a boost converter to the radio input.

When the radio fails to turn on and I take voltage measurements, I get 4.xV at the lithium cell, 1.1v at the charge controller output and about 1.1v at the output of the boost converter. 

Seems like the culprit is probably that super cheap charge control circuit I got as a set of 5 for barely any money.  I bought an Adafruit circuit that I've used in the past to reliably power Raspberry Pi projects in the past, but it's en route still.

I've considered that the radio may be damaged, but it works if I connect it to a 12v wall wart.  I just don't have what it takes to reliably make that power connection with my wall wart power supply to be 100% sure and run it through transmitting, programming and what not.  I've been using my fingers to press two wires to the + and - on the radio where the battery should hook up when I'm not using the lithium cell/charge controller/boost converter that I have rigged into the original battery case and that has been hit or miss. If I don't hold things JUST right, which I can't seem to do while I press the mic key, I lose connection.   

Since I was able to run the radio for a few hours when all the stars aligned and I didn't touch it, I don't think the radio is the problem.  I was able to program a single local HAM channel in, but by the time chatter started on that frq, the radio was not working.

In the mean time, I'll be home soon and I'll try to rig up something so I can connect my 12v source to the radio to try to eliminate that as the problem. 

[kjr18]

This charge controller that you're using, is it version with protection? You're talking about charge controller output, so I presume that it's the case, since simple TP4056 boards don't have any output pins, only input and battery. It looks like when transmitting it tries to take too much current, thus tripping overcurrent protection. Can you tell us what model is that Yaesu? It might help to find what current draw is expected.

[Nohippychicks]

The place I bought it from (Here's a link: https://www.amazon.com/gp/product/B01DRT4PWY/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1 claims, under "Specifications," that it does have protection but I didn't see anything more than that and blindly just assumed that because it's a knock off of some other design that it'd be OK.

I have bought a new Adafruit "PowerBoost 500," which I've used in the past and they have been reliable for a little over a year so far.

The radio model is an FTC-1143; I didn't even think of posting that!  Or maybe I thought I did and it just slipped.  I'll go with that one...

The battery pack model number is a Yaesu FNB-4, if that's needed as well. 

[KL27x]

I get 4.xV at the lithium cell, 1.1v at the charge controller output and about 1.1v at the output of the boost converter.

I am missing something if this makes sense to everyone else.

If you measure 4V at the lithium cell, how is the output of the charge controller 1.1V? Are they not one and the same, connected directly together, along with the input of the boost converter?

Maybe I have mistakenly assumed the charge controller is to charge the (li ion) cell/battery? Or is there more than 1 battery in your circuit?

Earlier you said this:

1.1V at the output of the charge circuit and 4.x volts at the input of the charge circuit

There was a charge circuit. Same as charge controller? Do you have it hooked up backward, by any chance? None of this makes sense.

It sounds like you are powering the charge circuit with the li ion cell. And then using the output of the charging circuit to power the boost converter? 

I feel like the kid that has to sit in the corner with a dunce cap. Cuz this sounds all kinds of stupid to me.

[KL27x]

No clarification, no answers from anyone else who is helping. OK, so this is my GUESS.

Lipo cell ->charge controller-> boost converter->12v load and I'd like some help.

Since other smart people are answering this like nothing's wrong, I feel like I am wrong.

But I would expect this to work a hell of a lot better:

Lipo cell > X > IN boost converter OUT > 12.8V output


External power supply > IN charging circuit OUT > X

The LiPo cell, charging circuit output, and the boost converter input are all connected together. (This is why there are two pairs of solder pads in the charge circuit board on the output side, marked as the Battery +- and the Output+-). The voltage you measure on any one of these 3 things will be the same. And it should be something like 3.5 to 4.2V, depending if the radio is on/off and how charged up the lipo is. You cannot measure 4V on the lipo and 1.1V on the charge circuit output unless you have it hooked up in some way I am curious to understand the purpose of. If you cannot draw up a schematic, maybe you can at least tell me what is the charge circuit? Why did you buy it, and what do you expect it to do? It's definitely not going to charge your lipo if you connect the battery to the INput and have no external power source in sight.

[edited for errors]
*Ok, just looked at your link to peek at this board. kjr18 suggests this board might be used for output protection. And looking at the board, he is correct. (But this doesn't change the way you connect it. Still gotta do it my way).

The big chip is the charge IC. The tiny SOT is a voltage detector. The medium chip is the output FET/switch.

The battery goes on the  B+- pads. It looks like the B+ and Out+ are in continuity. The FET looky thing is next to the ground pads, and it would be an NFET, drain on the OUT- and source on the B-. When the voltage goes under the specified level, the detector switches off the NFET and saves the day.

But you have soldered your battery to the Input pads. The input pads will likely be directly connected to the USB connector power and ground. That would be so you could hardwire a 5V power supply in lieu of using the USB connector. So if you plug it into a USB port in this state, and if the power source doesn't cut out immediately, you will damage your cell.

This explains your readings. You are powering the battery charger input with the battery and using the charger output to feed the boost converter. The IC detects no voltage on the battery and initiates trickle charge. This kicks the boost circuit on and gives the radio just enough juice to malfunction. And the resulting voltage at the output of the charger/boost input is 1.1V. Which is not enough to power the voltage detector, either.

[kjr18]

If you have your circuit like that one below, there are some possibilities why it has problems with working properly.  If your connections are different, then it might be the cause for your problems.

First problem: Your battery protection ic triggers because of too high current draw. Solution for this is to use different source.
Second problem: Your boost module might be not suited for lower input voltages, with battery voltage drop under load it might stop working.
Third problem: Your boost converter might be too weak to handle this task.

After all we might help you, but we need a bit more info about your circuit, used modules etc.



PS. God, I suck at Paint.

[Nohippychicks]

First problem: Your battery protection ic triggers because of too high current draw. Solution for this is to use different source.
...

I think this is the issue here! 

I'll take a picture of what I have when I get home, but that beautiful Paint rendition is pretty much the way I have it hooked up and those even look like the same charge circuits and boost converter I'm using, but it's tough to tell for realsies.

[kjr18]

I did some "tests" using a power supply, a boost converter and some load.

With 300mA load at 12.8V this crappy chinesium boost converter takes 1.3A at 4V.
With 500mA load at 12.8V it's 2.3A at 4V.
With 1A load at 12.8V it takes 3.7A at 4V. What a piece of crap, 60% efficiency.

You can use some notebook power supplies with some buck converters to drop voltage a bit, to desired level.

[KL27x]

^ But... 4kAh LiPo can output at least 40 amps without batting an eye. And possibly close to 200. Depending on what kind.

The discharge protection on the charge controller does not measure current, does it? It measures voltage, no? I think it doesn't care how much current is passing unless it causes a voltage drop below threshhold, which will be around 3V, min. Most popular cutout probably 2.9V and under. That size LiPo will sustain 3.3V+ easy at 40 amps. So the max current output cutout depends on the output impedance of the battery, which in this case is really really low.

I would suspect the boost circuit is the potential weak link, here. This can* cause the radio to not function. But this shouldn't create the 1.1V readings (unless there is some additional problem/malfunction/miswiring).

*I have no idea what the current draw of the radio is. But at any load, it shouldn't happen like what the OP described. If the output FET of the charge controller is tripping, there should more than likely be some other malfunction.

The 5/6 pin SOT  could be a dual voltage detector. It could theoretically sense the voltage drop across the output FET as a true max current shutdown. I'd be curious if it is wired up like that. OP could tell by looking at the traces coming off of it. Just looking at the pics, it does not look like this is the case.

Edit: Ok I didn't find a datasheet. But I found this video. I can't listen to the whole thing, but I scanned, and HE has a copy of the datasheet. At 1:08/9 is where you can get a peek that shows it. The FET is a dual FET. It lists voltage of the second detector to be 150mV. It probably measures the voltage drop across the output FET. How many amps that translates to, I don't know.



And it looks like thm_w figured all this out sometime last year.
I'll be busy removing foot from mouth.

[Nohippychicks]

Ok, I finally got home so I could piddle around with this thing again..  I attached a  picture of the little setup I have, and I'll edit this post if it doesn't show.



The white and red wires opposite of my hand are nothing, they were part of the old battery and attach to a little barrel jack.
The black wire swinging in the breeze is the negative out of the boost converter and is normally attached to a negative terminal on the other half of the plastic battery housing, I removed it for clarity.
Of course, the circuit stuck to the side of the battery is the boost converter and the other is the charge circuit, which I feel is the culpret.

I was able to get a semi-reliable connection to my meter and wall wart.  With the radio not programmed, I can't transmit even to test, so the most load I was able to get was by turning on the back light.  I read .066A with the radio on and squelch off and .101A with the backlight on.

I think tomorrow, or maybe Tuesday, my newer charge controller should arrive.  I should get a steady output from that one...and if that's the case and all this is fixed, I'll steer clear of buying the cheapest stuff I can get from now on, maybe.

Once again, thank you all for the help in this!

[KL27x]

I found the datasheet for the output protection IC. I wondered if/how one could increase the output current threshhold.

The discharge overcurrent detection voltage (VDIOV) is V6 when VDO is set to "L" after increasing V6 gradually from
the initial status 1.
2. 4 Load Short Circuit Detection Voltage (VSHORT)
The load short circuit detection voltage (VSHORT) is V6 when VDO is set to "L" after increasing V6 gradually after
setting S3 to ON from the initial status 1.
2. 5 Charge Overcurrent Detection Voltage (VCIOV)
The charge overcurrent detection voltage (VCIOV) is V6 when VCO is set to "L" after decreasing V6 gradually from
the initial status 1.
 

I think tomorrow, or maybe Tuesday, my newer charge controller should arrive

Good plan.

Just plugging in  a vanilla guesstimate of 100R tops for the Rds of the output FET, and overcurrent voltage detector of 150mV, that puts the max current in the ballpark of 1.5A, min. Perhaps up to something like 10A realistically, on the high end (unless it has charge pumps for the gate driver). This is just guesswork from what I understand in there. I just can't figure out that datasheet. It's a doozy. I suppose there's some oscillation of this circuit if that is what is creating the 1.1V.  It 'posedly has some hysteresis/smarts, but damned if I can read that schoo.

With 1A load at 12.8V it takes 3.7A at 4V. What a piece of crap, 60% efficiency.

I had to do a doubletake; I would be happy with those numbers on my own boost circuit. So with my calculator, I come up with 86.5% 

[Jwillis]

What is the model of the old Yaesu UHF transceiver.Without knowing the actual power factory rated consumption of your unit its difficult to know whether its a battery problem or a transceiver problem.On any transceiver TX will consume much much more power .A newer Yaesu FT-25R will consume 1.5 A at 7.5 volts on Tx where as Rx will consume 200mA and only 100mA on standby. Did you test the unit without the battery to determine the consumption at standby ,Rx and Tx .Rx may be difficult unless you have signal traffic on a second transmitter on hand. If your transceiver if dropping the battery that far down so quickly I would say theirs a problem with the unit its self. OR Some LiPo packs have over current protection built in .Meaning they will shut down if the current draw is to much.Its there to prevent over heating of the battery pack during a possible short . Older NiCads and NiMh packs don't have over current protection .If your transceiver is pulling more current then what the LiPo pack is rated for it won't work.
What is the charge controller circuit configuration in relation to the battery and transceiver.A simple drawing will do.Depending on how your charger is wired you could be over working the circuit and its shutting down.
You can test with any 12 volt supply.A lead acid battery will do.But find out what the rated factory current consumption At TX for your Yaesu first. Put a fuse rated for or less than that current consumption in series from the battery before your amp meter. Its the simplest means to control current without calculating a suitable high Watt resister . Standby shouldn't consume much.Rx a little more and short burst of TX will consume much more.  Direct battery to your unit will turn it into a toaster if theirs a problem with the unit itself  so use a fuse to limit current.
I work on a lot of mobiles at work and a lot of them come in with the back ends burned out.This is a because the antennas get broken off and are continued to be used.Then they get turned in for repair only because they start to burn out fuses continuously.Not to say this is your problem I'm just venting.
More information  on you transceiver would be nice.You should be able to find a copy of the operators manual online very easily.

   

 

[Nohippychicks]

What is the model of the old Yaesu UHF transceiver.

This is an FTC-1143.  If you can find a manual, please feel free to link it here.  The best I can find is "How can I program this radio,"  type threads in various forums.

I can't transmit with it at all, as I can't get it to reliably turn on, so I haven't programed even a single frq in.  Even NOAA hasn't been tuned in on this handset.

Since you work on these things, maybe you have more information, though!  When I key PTT, I get "ERROR," in the display.  Do you know if that's what it should display with no frq to transmit on? It seems reasonable.

When you say that these get burned out, what should I look for to check for that?  I , personally, have never tried to transmit without the antenna installed, but the history isn't known.  I've also not opened the radio, figuring that if it turned on and had a display that would be a good enough function check for the time being. 

[TERRA Operative]

I would start by replacing that charge/protection module with a charge only version, your battery already has a protection circuit in place.

The protection circuit in the charge/protection module maaayyy be causing the issue as it isn't designed for your battery in particular.

[Jwillis]

I sent a request direct to Yaesu for specifications but will be a few days for a reply if at all.Sometimes I get lucky when I ask nicely.
The error your getting is most likely not enough amps are being supplied.Our mobiles will do the same thing.That tiny buck boost will deliver 12 volts but the available current will drop off significantly.An older model like yours will likely draw more current than a newer model.So I wouldn't doubt if it goes as high as 2 amps or even 3 amps on Tx (PTT). My personal mobile will draw 20 amps on a transmit .Just popping in a FET to bypass current probably won't work because your dealing with 2 separate voltages.12 volts at the gate and 3.7 at the source.3 of your LiPo packs wired in series might not even be enough.But its worth a try since theirs 4 amps available at 11.1 volts in series.
Hand held s were not designed to be used to be in constant Tx. Even the best ones will only function for a short time.

[kjr18]

I had to do a doubletake; I would be happy with those numbers on my own boost circuit. So with my calculator, I come up with 86.5% 

You might be right. I should've used a dmm instead relying on my load led display.