Panasonic cordless phone shows charging but actually it's discharging

There are lots of cordless phones that meet the DECT specification, all different, especially in ways that are possibly outside of the DECT spec. such as the battery charging circuit. I'm an electronic design engineer and very experienced in the field. I got so sick of my Panasonic cordless phones going through NiMH batteries every couple of years when, if charged properly, they should last much longer, (I fitted a 10 cell pack of Panasonic AA NiMH batteries, (date coded 0405), as 12V backup in my burglar alarm system, maintained by a precision CVCI float charger and on load testing them, they're still good after nearly 19 years service).

From my investigations I suspect that the Panasonic system is a switch on/switch off bulk charging system which continually unnecessarily cycles the batteries and thus dramatically reduces their life. I suspect this is compounded by too high a 'charged' voltage detection that exacerbates the problem by overcharging the cells. To solve both of these problems needs a two-stage solution:

1) Firstly to install a system compatible precision float charger which can maintain the battery level at a constant 'fully charged' level by strict control of the charger current and a precise limit on the upper battery voltage to eliminate over charging.

2) This would then make the crude Panasonic 'switch on','switch off' charging system redundant and it can be disabled to the 'charging' condition to fully implement a constant voltage/constant current (CVCI) float charger that does not continually cycle the cells.

The following mod. implements part 1) of the solution and eliminates any over-charging with a system compatible charger solution. After a period of assessment, part 2) can be investigated to eliminate any residual unnecessary battery cycling.

So, back to my Panasonic DECT cordless phones, (KX-TCD220E):

When I took the cradle of a slave handset apart I was surprised to find that the charger circuit is very crude and possibly at least partially the cause of the prematurely dying NiMH batteries - it consists of a 6V AC 0.5A transformer + rectifier + smoothing/reservoir capacitor in the wall adaptor, and all that's in the cradle is a couple of 22 Ohm parallel connected SMD resistors in series with a Schottky blocking diode straight to the cradle battery contacts & that's it. In other words just an 11 Ohm resistor straight to the handset batteries via a Schottky blocking diode. I wondered why the batteries got rather warm to the touch & all my phones are equally problematic. I couldn't believe a firm like Panasonic would be guilty of such a bodge, and certainly in the handset there appears to be some sort of control on one of the input battery connections, but what algorithm it follows I don't know, but from all the problems it undoubtedly does seem to overcharge the batteries. Oh, and BTW, beware of the polarity - on mine, Panasonic have managed to opt for -ve to the centre connector making it out of step with most other adaptors...

Anyway, to cut a long story short it was clear that nothing could be done to the Panasonic charge control for the moment, so I turned my attention to the charger itself and I bought some nice very cheap CVCI adjustable voltage buck switching regulator modules for next to nothing from China, (see images), which would produce near enough the correct charging parameters for NiMH batteries and incorporated them in the slave and base station cradles and carefully adjusted them, (in place of the existing charger 22 Ohm dropper resistors which should be shorted out when the mod. is complete), to fall back to a very low maintenance charge once a full charge voltage of 2.86 volts for the series AAA battery pair had been reached, (i.e. when the batteries are fully charged), so now there's no possibility of battery cooking as the module efficiently limits the charger voltage and current available to safe levels. The whole system still charges, indicates and works 100% normally - the additional module simply removes any possibility of overcharge by replacing the 22 Ohm dropper resistors with a CVCI charger module. The handsets are untouched and remain DECT compatible. If you do this modification I'd advise changing the batteries at the same time as the existing ones will not be in good condition.

Nothing kills ANY battery quicker than pouring full charge current into them when they are fully charged which is what seemed to be happening, and , of course, all batteries have a limit on the number of charge/discharge cycles they can achieve. The next step will be to investigate disabling the Panasonic charger control to see what effect that will have on the system indications.

PS. Just to emphasise, this step 1) is designed to allow any existing Panasonic battery charge control to do its thing, (whatever that is in my particular KX-TCD220E setup - let me know if you have any idea), but to prevent it overcharging the battery unnecessarily by carefully regulating the available charger output only and it seems to work perfectly on my two base units and six handsets. Only time will tell if it's effective or not on its own as the system can possibly still cycle the batteries. I've come across many cases like this based on the misconception that NiMH and NiCd batteries can be safely charged with a constant current irrespective of the state of charge which simply isn't true - all battery chemistries will be damaged by unnecessary overcharge, (although some are more sensitive than others...).

I've now had a look at an online workshop manual/handset schematic and nowhere can I find the voltage that the handset cuts off the charge to the handset batteries, just at what voltage it initiates the charge, but as far as I can see, the suggested modification should work OK with the Panasonic firmware control, which becomes essentially redundant as a result of this mod.

PPS. All of the Schottky blocking diodes in my chargers were OK, but there is no harm in upgrading them to 3 Amp. devices, apart from a possible minor increase in reverse leakage.

Panasonic Cordless Phone System type KX-TCD220E Slave Handset Charger Modifications.

 The following modification is specific to my cordless phone set, (KX-TCD220E), but should be relevant for other Panasonic variants. Look at all the complaints of these phones unnecessarily killing re-chargeable NiMH batteries and know that big names don’t always get it right and Panasonic are no exception. This should be suitable for someone with reasonable knowledge and soldering skills.

Existing Slave Charger circuit                                                                   Variable CVCI* module (Aliexpress)

 Modified Slave Cradle Charger Assembly (See separate instruction for base unit modification):  

1.       Prepare the CVCI control module by bridging the ‘5V’ and ‘Adj’ voltage setting points on the rear of the module pcb with a globule of solder as shown. This will give optimum fine control of the module output voltage to 3.2 Volts by adjustment of the miniature CVCI unit voltage adjustment potentiometer. See setting up instructions and module photographs.

2.       Unplug the DC plug from the cradle assembly socket and unscrew the two cradle assembly securing screws. Separate the two parts of the cradle assembly. Insulate rear of power socket J1 with insulating tape patch to prevent any contact with the charger module.

3.       Break the pcb connection to the Schottky diode anode by cutting a small section out of the copper track immediately behind the diode anode. Unclip the PCB from the cradle. Solder three c. 4” insulated wires to module connections Vo+, In+ and Gnd and mount module on sticky pad in any clear position. (See photograph for suggested position).

4.       Connect the CVCI module Vin+/In+ connection to the cradle power socket +ve (outer) connection.

5.       Connect the CVCI module Vo+ connection to the isolated Schottky diode anode connection.

6.       Connect the CVCI module GND connection to the cradle power socket negative (centre pin) connection. (The module ‘EN’ connection is left unconnected). Clip the PCB back into the cradle.

7.       Important! - By-pass the 22R resistors on the cradle board by connection a wire between the Schottky diode cathode connection and the cradle PCB +ve handset contact, (or simply connect a short piece of wire across one of the 22R parallel connected resistors).

Optional ‘Power On’ indication upgrade: Drill a 5mm hole in the charger cradle midway between the centre of the ‘Panasonic’ legend and the bottom of the cradle and insert a 5mm green LED, secured  on the inside by a blob of hot-melt adhesive. Wire the LED anode via a 100 Ohm ¼ W resistor to the Schottky diode anode/CVCI module Vo+, (not the handset contacts) and the cathode end to the power socket –ve.

Setting up (Must be accurate for best results). 

1.       Check all connections, remove any handset and connect a temporary >2 watt 10 Ohm ‘dummy load’ resistor across the two external battery connector spring terminals and plug the adaptor DC plug into the cradle socket.

2.       Connect an accurate digital test meter across the above load resistor, power up the system and adjust the output control of the CVCI module to give an exact indicated voltage of 3.2 volts across the 10 Ohm dummy load resistor.

3.       Remove the dummy load resistor, reassemble the cradle and that’s it! Don’t forget to replace the old AAA batteries – I suggest Ansmann 1100mAh AAA rechargeable NiMh batteries for best performance.

N.B. – Always leave the handset sitting in its powered charger when not in use. Following this modification the batteries will charge as fast or faster than before, but will no longer over-charge, the handset will run cooler, (especially as the modification is much more efficient than the crude ‘dropper resistor’ original system), and the batteries should last much longer.

The modification can also be applied to the Base Unit as follows, even though its power circuit is more complex: 

Panasonic Cordless Phone System type KX-TCD220E Base Unit Charger Modifications. 

1.       Prepare the CVCI control module exactly as in 1 above, including soldering the three wire tails, but in this case, solder a 1A 40V Schottky diode, (e.g. 1N5819), in the Vo+ pcb hole, (anode to the pcb), and connect the Vo+ tail to the diode cathode instead of directly to the board.

2.       Unplug the base unit DC power and landline plugs & remove the four case screws on the underside.

3.       Unscrew the screw securing the little LED sub-board. This allows the main board to be tilted back for access.

4.       Mount the CVCI module on a little sticky pad in a suitable position as before, (see photograph for suggestion, but anywhere that clears the case will do).

5.       Unsolder the existing battery +ve wire, (Red in my unit), from the spring battery connector assembly and connect it to the Vin+ tail from the CVCI module – insulate the connection, (which goes to the +ve supply on the main board), with a piece of silicone or heatshrink sleeving. Alternatively the existing red connection can be unsoldered from the main board + connection and a new connection made from the main board + connection to the module Vin+.

6.       Connect the module Vo+ connection, (via the 1N5819 diode), to the +ve battery spring terminal on the battery connector assembly. Solder a shorting link across one of the battery charging resistors on the topside of the main board, (i.e. R3 – marked 220 – see photograph below).

If the optional LED ‘Power On’ modification is required then the 5mm hole can be drilled in the case in the same place as above and the LED should be connected with its cathode to the battery assembly –ve connector and the LED anode connected via a 100 Ohm resistor to the Module Vo+ output connection on its pcb, (not the Schottky diode cathode). See pictures below.

The set up is the same procedure as above, (3.2 Volts), with the temporary 10 Ohm 2W resistor connected across the battery spring cradle output terminals, (with no handset installed) and once done, remove the 10 Ohm temporary dummy load, replace the main board, the LED sub-board and the rear cover and don’t forget to replace the handset batteries with new ones!

As a final check that the modification has been done correctly, secure two fine temporary test leads under the battery pair connections in the handset, (charger contacts end), place the handset in the charger cradle and wait for the charging indication to cease. The measured battery voltage on the test leads should then be between 2.86V and 2.88V with new batteries. If it is significantly different then check that you have done the modification and the initial setup correctly. A minor discrepancy can be trimmed out by a small adjustment of the CVCI board pot..

*CVCI module = Constant Voltage/Constant Current very high efficiency precision switch mode voltage buck converter with current limit. Available from Ali Express: https://www.aliexpress.com/item/32801569...

Well it's a couple of months or so since I modified my two base units and six handsets as above and not a single problem to date. It will take several years to prove the effect on battery life, but as the CVCI modules limit the cell charging/running voltage to 1.43 volts per cell, they should last for many years longer than they used to.

All slave and base unit handsets continue to function perfectly as normal.

Have a slightly different problem. 1 out of 4 handset is forever charging even though battery is full. I can put a known full battery (shows full in other handsets) into this one and it will show 1 bar. If the handset charging circuit think battery is not full. All of this without the charging station. The charging station is fine as I can put other good handset in there and battery charing behavior is correct (shows charge complete when full)

So obviously if the handset thinks a full battery is not full, then its always charging and generating heat like what I am seeing. Going to probe the circuit board (pic below) for voltage readings. Looks like there are diode on the battery side also. Wondering if others have had this problem.

I have 4 handsets and using a known good fully charged battery, I get 2 handsets showing 1/3 bars, 1 showing 2/3 bars, and only 1 handset showing 3/3 bars.

Dug a bit deeper into the schematic (didn’t find exact but perhaps close) Battery voltage is fed to the IC (likely pin 58 of basedband IC SC14430). IC reads voltage and make decision to trigger the external charge circuit (transistors + diode system). 3 Voltage islands on the handset board (3.3V, 2.5V, 1.8V) identical readings between good handset and bad one.

On the good handset, had the handset opened up with alligator clips on battery post for 2xAAA power supply. The alligator clips accidentally touched the 2 charge terminals as its easy to do with the back cover off. Battery and charger terminals next to each other are reverse polarity and the good handset also shows 1/3 bar on full battery after touching. There was no spark or noise. Something must be really easy to damage with such low voltage and current.

Remove the single screw from the base unit. Pry the base open. Inside there is a circuit board, on it is a small black component (a magnifying glass will help) there is a small diode symbol on the board next to it. Using a soldering iron, melt the solder holding the diode and remove it from the board. Get a new diode (ebay?) and solder it in place of the old one ( it must go the correct way round). It doesn't have to look pretty, so long as the connections are ok. - That is it.

I charged a set and put them in the failing phones. They also said they were charging. But slowly, they went dead because ‘charging’ is a false message.

I want to run this to ground when I can free up some time. I’m pretty sure I’ll find a charging/supervisory chip that’s gone bad, so the batteries don’t actually get charged. Probably the incoming voltage from the charging base goes to a microprocessor pin that calls it ‘good’. Dorks! They designed it wrong! And... it’s probably going to be something the average Joe can’t repair.

Maybe I can go into the Panasonic Wireless Phone repair business, since I retired from years of electronic design engineering. If Panasonic won’t fix it, maybe I can! They should pay me!

We’ll see. Don’t hold your breath. :-)

Dan Smith

hi,

i have a somewhat similar problem with my panasonic handset.

my handset is 110v and i live in a country with 220v. i usually just get an adapter 220v to 5v and splice them up. this time i accidentally switched the wires while splicing.

when i put the phone on the dock, it somewhat reversed the phone. meaning when i remove it from the dock. it says that its charging and i cant use it. and when i put it on the dock it thinks that it is out of the dock and i can use the phone.

i tried the dock again with another phone, and it did the same. it somehow reversed the phone. after that i doubled checked my splice and found out the i reversed the wires.

is there a way to fix this? not sure if the polarity is the issue. for now i cant use the phone.

thanks in advance

There are 3 possibilities - bad batteries, bad charging base, bad contacts. It's clear that bad charging base (defective diode) has happened to a number of people and is probably the main reason people come here (after all, you will have replaced the batteries before you get to this topic). Why not just close this topic - it's been solved!

The question has not been fully answered. People who want a thread closed ought to have their own circuits examined. In the case of my phone, the problem is neither the base nor the batteries nor the contacts, it's the charging circuit in the handset. I know this because I cleaned the contacts on the handset, transferred its batteries to a different phone, and put that phone into the same base. Phone #2 charged fine in that base.

instead of using NiMH, I'm trying to use rechargeable lithium ion battery.

Seems like they used a TRANSISTOR to stepdown the 2.4~2.6V input (paralleled by a 2.7v Zener) to 1.8v.

I have no clue what the TRANSISTOR or the zener are capable of(no spec), but the 6v charging voltage tells me that I can probably connect lithium battery to the battery terminal and let the zener take care the rest?

anyone has any thought?