Threaded Mode | Linear Mode

albertofenini · 04-06-2020, 12:17 PM

Hello everyone, first of all we hope that everyone is doing fine during these hard times !
Best wishes for everyone and good luck to all of us !!

Yesterday we tried to fix an HP82004A, a Power Reserve Pack for the Classic serie.

We use these to keep a Classic Battery Pack always ready, and while we normally
connect it to a Classic Power Supply, this one was connected with a USB cable to
a Kindle USB power supply.

Worked fine for a while unless the led that usually turns on when a battery pack is in place went off.

Yesterday we replaced the led and to our surprise it didn't fix the situation, as a
matter of fact the original led was working.

We replaced then the transistor with a spare unit like the original, and what's happening now is
that the led stays on even when a battery pack is not in place, just because a Power Supply is connected to the Reserve Pack.

We checked the voltage and the current using another working Power Reserve Pack
(which behave normally) and both provide a 15 volts at the terminals and if a battery is in place both provide a 55 mA current.

That makes us think that the repaired unit is working, but then why the led is always on when
the Power Supply is connected ?

Thanks for any idea you want to share !!

Stay safe everyone

Pictures here : Repair of an HP82004A Power Pack

TomC · 04-06-2020, 03:32 PM

Hello:

I too have wondered about the internal circuitry of the Battery holder - I'm surprised there does not seem to be a schematic online. Given time, I may sketch one out.

During normal operation, the LED illuminates only when the battery pack is inserted. There is no mechanical switch, so I believe the LED is illuminated when the battery pack is being charged. I hope that someone knows the cross reference for the transistor in that package.

Regards,
TomC

albertofenini · 04-06-2020, 03:44 PM

I believe there are some schematics, but it's a fairly simple one I may draw it and share it,
we used exactly the same transistor, the 4-094 is used in the PSU of most of the
Classic units and we had a spare, although with an HFE of 14, which we believe good enough.

This why we are surprised that id does not behave like before having substituted the
transistor with an original one and not an equivalent.

Also, even if not shown in the pictures, even the original led has been re-utilized
given that it was working.

Besides, once the same voltage is provided and the same current can be drained
it should be working right ?

Let's see if someone has a hint ....

AndiGer · (This post was last modified: 04-06-2020 04:00 PM by AndiGer.)

Maybe this helps?

grsbanks · 04-06-2020, 09:34 PM

Not sure why that guy is so worried about charging NiMH cells too quickly. They can take charging currents far above what NiCd cells can take. Typical AA cells can take 500mA no problem.

teenix · 04-06-2020, 11:41 PM

Hi all,

I hope everyone is well and keeping busy if in lockdown.

If the circuit described in the link above is accurate then there must be a short somewhere in the electronics or plug, otherwise it would be impossible for the LED to light.

Disconnect the end of the 100 ohm resistor to the plug. If the LED is still on, the LED anode might be shorted to the + battery wire.

The circuit is simple, keep disconnecting/reconnecting bits of it to determine where the fault is.

As the electronics is assembled as a rats nest, it would be easy for something to short.

cheers

Tony

ijabbott · 04-07-2020, 06:37 AM

(04-06-2020 09:34 PM)grsbanks Wrote: Not sure why that guy is so worried about charging NiMH cells too quickly. They can take charging currents far above what NiCd cells can take. Typical AA cells can take 500mA no problem.

They are much less tolerant of being overcharged than NiCd cells though.

albertofenini · 04-07-2020, 06:45 AM

Thanks Tony

we'll try to rebuild it from scratch and will see what's wrong,
however, giving that it's capable of 50mA like the other one can we assume it is charging a battery ?

Take care

(04-06-2020 11:41 PM)teenix Wrote: Hi all,

I hope everyone is well and keeping busy if in lockdown.

If the circuit described in the link above is accurate then there must be a short somewhere in the electronics or plug, otherwise it would be impossible for the LED to light.

Disconnect the end of the 100 ohm resistor to the plug. If the LED is still on, the LED anode might be shorted to the + battery wire.

The circuit is simple, keep disconnecting/reconnecting bits of it to determine where the fault is.

As the electronics is assembled as a rats nest, it would be easy for something to short.

cheers

Tony

albertofenini · 04-07-2020, 06:46 AM

yes it does ! Thank you so much !

(04-06-2020 03:59 PM)AndiGer Wrote: Maybe this helps?

grsbanks · 04-07-2020, 06:47 AM

(04-07-2020 06:37 AM)ijabbott Wrote: They are much less tolerant of being overcharged than NiCd cells though.

Very true, but there's little risk of that happening while charging them at 50mA!

albertofenini · 04-08-2020, 01:37 PM

I'm confused ...
If I look at the schematic provided by AndiGer the transistor 094 is replaced with a 2N3904
If I look at old posts or the Jaques Laporte documentation the 094 is replaced with a PN3646
Which one is correct ? Or both are possible replacement ?
On Wikipedia a suggestion for replacing the 2N3904 is the BC548, does anone has ever tried this ?

take care !

teenix · 04-08-2020, 09:55 PM

Hi Alberto

I think this is a circuit of the charger. (From Tony Duel)

cheers

Tony

albertofenini · 04-08-2020, 10:03 PM

Thanks Tony

What do you think of what I have written for the transistors equivalents?
Take care !!

Dave Frederickson · 04-08-2020, 10:32 PM

(04-08-2020 01:37 PM)albertofenini Wrote: I'm confused ...
If I look at the schematic provided by AndiGer the transistor 094 is replaced with a 2N3904
If I look at old posts or the Jaques Laporte documentation the 094 is replaced with a PN3646

According to the HP Master XRef, an 1854-0094 transistor crosses over to an ON Semi MPS3646.

Coincidentally, Sphere has an 1854-0094 in stock.

albertofenini · 04-08-2020, 10:38 PM

Thanks Dave
What do you think of the use of a BC548 as a replacement?
Thanks and take care !!

teenix · (This post was last modified: 04-08-2020 11:33 PM by teenix.)

(04-08-2020 10:38 PM)albertofenini Wrote: Thanks Dave
What do you think of the use of a BC548 as a replacement?
Thanks and take care !!

The BC548 has a bit more gain, but should work.

The pinout for the mps3646 is [e b c] looking down from top, flat to bottom.
The BC548 is [c b e].

If the LED works as expected, hook the battery up afterwards and measure the current into the battery.

cheers

Tony

Dave Frederickson · 04-09-2020, 12:01 AM

(04-08-2020 11:32 PM)teenix Wrote: The pinout for the mps3646 is [e b c] looking down from top, flat to bottom.
The BC548 is [c b e].

The illustration on p.2 of the datasheet is wrong. See p.5.

Dave

albertofenini · 04-13-2020, 08:08 PM

Hello everyone, we fixed it !
Thanks to everyone’s contributions !
We used a BC548 as a replacement, hfe was over 200
We wired the pieces differently and isolating the transistor legs just to be sure
Now the led only turns on when a battery is inserted
Voltage provided is 15,3 volts and current draining is about 50 mA
Although differently everything fitted in the narrow space available around the power connector
Thank everyone !!!

Pictures updated here : Repair of an HP82004A (updated April 13th 2020)