Battery Tender

[jheatac]

My next project is a battery tender. SEM-1562A by Schumacher

Again - this is about learning. A new unit isn't but $30 so if I really needed one, I'd just go get another.

Two things - First, even when idle and the unit appears off, the unit has a high pitched whine from I assume the SMPS. I have had several of these units and it isn't a normal sound. I've heard this on other devices too and associate it with a bad capacitor but that isn't based on anything. I do have pretty sensitive ears. It seems to quiet somewhat but not go away under load.

The reason I took this apart is that after a time, it goes black instead of lighting the green charged LED. I feel like it charges 100% and shuts off, but I am trying to confirm this as I know it is charging.

My first question is if this is an acceptable waveform for the high side of the transformer or if this is indicative of something. I could not get a steady waveform and the measured frequency was jumping around. I don't recall so much of a ring on other SMPS i've played with. By reading the scale of the scope, it looks like it is at 50khz, correct?

I did replace a 1AM SOT23 transistor on the green LED, but I am thinking I blew that during my testing. I've got it charging a battery right now so I'll know shortly.

I might add that the unit does put out the rated amperage. It is a 1.5a unit and I loaded it and it put out 1.6a

Thank you!

[Smokey]

Battery tender... that's the new dating app for batteries right?

[coppercone2]

its actually a robot food

[floobydust]

Post pics of the board.  "Battery Tender" is a brand name not affiliated with Schumacher.
These car battery float chargers are notorious for using cheap electrolytic capacitors. The small one typ. 22uF 50V for the SMPS IC is critical, and it might be fading.
The caps on the secondary will heat up when load is high and that is when they usually cause trouble. I would test the capacitors, first careful to make sure they are safely discharged, look for bloat or a bulged top.
Your primary waveform looks OK. The frequency jitter is intentional to limit RFI, the IC's have this feature.

edit: Schumacher SEM-1562A analyzed on another forum is a simple linear power supply. If they updated it to SMPS then the model number you'd think is different?

[jheatac]

Here are board pictures

[JustMeHere]

SMPS can produce audible frequencies when the current draw on them is very low.  They switch from PWM to PFM

[floobydust]

The black arc spot on C8's pin does worry me. What happened to this charger?
It's prob. under warranty PCB date code is last year.

[jheatac]

I don't know that anything outside happened to it.

I do not see a spot around C8.

There is a slightly dark trace when I first dismantled. Top left mounting hold - two traces to the right (or left of R19a). Specifically the underside of D4 rectifier diode. I've inspected it a couple times and don't see any damage or de lamination, it just looks a little dark.

There is a slight discoloration on the open pin of the transformer near R35. I accidentally arced it to one of the adjacent pins when testing. It doesn't appear to have caused damage.

Ok so I kept looking and see what you are referring to. It appears to dried rosin or at least something that resembles that.

[Buriedcode]

The ringing in that waveform is typical of a flyback switch-off period - damped oscillation between the MOSFETs D-S capacitance and the primary of the transformer.  As to whether this unit *should* have that waveform, I don't know - it depends on the controller and if it's operating in DCM or not.

The battery chargers I've seen - I'm assuming that what it is, as "tender" doesnt' mean much to me - are just flyback power supplies with some comparators/opamps controlling the feedback  to provide either constant current to the battery, or, when bulk charge is complete, constant voltage.  It then waits for the current draw to drop before switching to "float" voltage.  This can all be done in analogue, although some have a micro there to do extra things like pulse charging.

I would monitor the swithcing waveform when a battery is connected - and then monitor the output current and/or voltage.

As a side note, the last SLA charger I repaired also had a whine and a flashing "charged" LED - turned out to be a broken feedback winding in the transformer, because it wasn't sufficiently attached to the board, so years of being bumped around with only the winding solder joints holding it to the board made one give out.  This meant the primary switcher wasn't getting its power supply, so was just hicupping off the start circuit.

For safety reasons, I would not probe the primary side during operation to start with - you can often diagnose or at least narrow down the problem just by looking out the output of the output diode. (that TO220 on the bottom right heatsink).  Typically things that fail are the mechanial first - connectors, switches etc.. - then the power components, switching elements, diodes, bulk caps.

[floobydust]

This is the spot I thought looked odd. OK then, no metal shavings, no water drama, no bug crawling around inside etc.
The usual I do is a visual, look for fractured solder joints on the through-hole parts, take an old toothbrush and brush off solder balls, IPA and Q-tip if the board looks dirty.

When the unit seems to shutoff/die I would measure the secondary DC voltages to see if power is still there.
It needs to have say 12-15V for output but also regulated power for the STM8 MCU and I don't see its Vreg. Maybe that zener ZD1.
What markings on on U6, U1? Do you hear the output relay click on/off?

These can be sophisticated with the firmware in the MCU. Some smart chargers will not output anything until they first see many volts at the output. It might even just trip on high end-of-charge voltage as a fault state, but the SMPS squeal I would check power and the caps first.

[jheatac]

OK - so dumb question first, then I'll back it up with why.

Is there any reason/way that removing the Switching Controller IC (U1) from the circuit would cause a short circuit? It is an AP3013A

So in taking measurements on this, I let the smoke out of it by shorting the multimeter lead across two pads. I've learned that lesson and have moved to leads with less exposed metal at the end.

Charbroiled R19a, R19b. Blew the fuse, R15, R5a, and the Mosfet.

I replaced everything and powered up and U1 wasn't working. I only had 0.8v at the VCC.

In measuring all the resistors feeding it, I had come to the conclusion (right or wrong) that the chip must be shorted, pulling down VCC and was going to remove it to see if the voltage returned.

I can't imagine removing it from circuit would cause a short, but my best guess is I floated something near it with the hot air in removing it, causing a short. Lesson two learned - be sure to check everything around after using hot air.

For anyone joining in without reading the whole thread. Yes, I'd like to fix this for the satisfaction, but it is really just hands on learning for me. It has been good practice soldering 1208 and 0805 resistors, and a SOT 23 on the secondary side of the board.