Piaggio X9 Evolution 250 scooter incorrect mileage odometer reading

A Piaggio X9 scooter that had electrical problems after being in the rain.

Water had got into the clocks and the electrics were not happy, many dash lights randomly lit.

After drying out, the scooter was all ok except that the distance / mileage was reading 99998 km or 62498 m (bike had been set to miles but had reverted to km).

This was not good ! service history and re sale is hurt with the mileage so wrong, also it wouldnt increment ! so basically broken.

A google showed its not uncommon, water gets in at the edges of the push buttons, I found a forum post that recommended sealing the buttons and putting bags of moisture absorbing gel inside the clocks.

Surprise surprise the clocks were opened to find that the buttons had been sealed and little bags of gel put inside already by whoever had the same problem last !

The sealent had gone off / was not the correct type and had let water back in.

I also found people on the web talking about the mileage going wrong, reading 99998. I figured the mileage had got corrupted.

The mileage is stored in a serail eprom 24c02

I removed the chip and read its contents, the chip read fine and comparing the contents to a good dump I found on a greek forum, it looked valid.

This was the first thing that made me think that the chip and contents were fine, just that maybe the MCU could not read it at boot up.

Here is how I powere the clocks on the bench.

I checked that the serial eprom was powering up by checking continuity to the supply rails, they were ok.

I checked continuity of the data pin to the MCU,this was present.

I checked continuity of the serial data clock pin to the MCU and could find no connection, this did not seem right !

The data pin also connected to a large pad on the edge of the pcb, presumably for factory programming, sure enough the clock pin did not go to any of these pads, however one of the MCU pins did.

I fitted a link from the eprom clock pin to this pad and.....

Result, the correct original mileage now reading from the eprom at power on.

It looks like the link for this track in the multi layer PCB had gone.

The clocks were re sealed and re fitted to the bike, all reported to work ok, mileage reading, incrementing and storing ok.

Grundfos watermill shower pump transformer faulty. SS45DLV.

A transformer from a shower pump, no output.

Opened it up and check of the windings showed the input was open circuit. Close inspection showed a thermal fuse taped to the winding, it was open circuit.

A new fuse was selected, original was 150, replacement to hand was 135, I was confident this would be fine and the cause of the fuse going was age related fatigue and not a genuine fail. The transformer smelt fine and ran fine afterwards. The new fuse was taped back onto the windings with transformer tape.

The input selector switch routed the mains to one of three tappings. I measured the ac output at the three settings, from memory they were ~30vac, 35vac and 40vac (possibly the 45 in the model number is 45 volt max ?).

Hitachi VNR10C209 LCD backlight inverter faulty.

A faulty backlight in an industrial lcd touchscreen.

The inverter board was a seperate PCB, a quick check of the thermal fuses showed that one had gone. A nice easy fix but not an immediately obvious one, I had not seen an inverter with this protection before.

In the picture you can see the top fuse I had replaced and the bottom one is the factory fit.

Both replaced, the lower one is not bonded to the transistor yet.

I stuck them down with silicon sealant (weighted down until set), fine at these temperatures.

The fuses were 115 degrees. On soak test the chips it was bonded to never got above 40 degrees so I am putting this one down to age fatigue on the thermal fuse, not uncommon in my experiance.

Alinco DJ-193 no sound from internal speaker.

An Alinco DJ-193 where the built in speaker no longer worked, headphones worked fine.

Opnening was fairly straigtforward, (after finding nothing on the web).

Remove the antenna and pull of the rotary knob.

Then just flex the sides at the points shown to disengage the metal tabs from slots in the case and tilt the whole metal lump out.

A nice straightforward one, the fault was clear to see immediately, the speaker wire had at some point been pinched between the case and the metal assembly and been cut clean off. It had never been opened before so presumable from assembly at manufacture or the wire had moved through use ?

The wire was re attached and the speaker worked again.

LG 32LX2R Faulty power supply.

A LG 32" LCD TV that would not power on, the fault occurred following a mains power outage.

The PSU board looked fine, no bulging or stained capacitors. Aquick google went straight to the point, somebody with identical experiences who recommended changing the output stage capacitors (circled) anyway.

http://forums.moneysavingexpert.com/showthread.php?p=27092505

post number #31

The capacitors were sourced, 14 in all. The board was of reasonable quality, no problems removing the old ones.

Here is a close up, all the old capacitors looking ok

So in went the caps. I measured all the old capacitors and they were all in spec for capacity on a basic Fluke multimeter....

However the TV now worked. I would be interested to know the failure mechanism of the capacitors but a good result and Google delivered as the capacitors really dint look suspicious.

Update:

Here is the capacitor list:

2200uf 10v x 1
1000uf 35v x 4
1000uf 25v x 2
1000uf 10v x 3
680uf   25v  x 2
100uf   50v  x 2

Update:

The likely failure for the capacitors is poor ESR, (equivelant series resistance) at given frequencies. I have access to a proper LCR bridge. A quick measure shows ESR varies with physical size on new capacitors, I will need to look into specs for given values and see if there is a large deviation from spec.

Update:

Ok, I had kept the capacitors and measured them all. Here is a table of typical ESR values. This is from a manual of an ESR meter made by Peak electronics. I hope they dont mind me using it, here is a link to there meter http://www.peakelec.co.uk/acatalog/jz_esr60.html

There are three 10v 1000uf capacitors in the TV PSU, two measured just under 0.2ohms but one measured 2.1 ohms, so it looks like this was the likely culprit.

The esr meter above can be used in circuit, I hope to try one out soon !

Dell 1708fp ultrasharp backlight fault

A dell 1708 monitor, the monitor would power on, screen would display for two seconds then the backlight would go off. You could faintly still see the graphics in the background.

This was exactly the same as the Samsung 540n I fixed in an earlier post, that was bad PSU caps so I was expecting to see similar inside.

Disassembly was straightforward. Four screws in the middle of the back,( behind the stand bracket), then unclip the bezel all the way around (this was tight !). Pull the back off, remove two screws from each side to release the electronics assembly from the panel, remove back-light wiring covers then cables, remove main screen cable. Finally to get to the component side of the board, remove output connector locking screw posts then screws from the back of the PCB.

I wasn't disappointed, more bulging brown stained caps than you could shake a stick at, six in all, all clearly failed.

They should be 35v 330uf, removed they were measuring about 150uf.

Here they are in all their glory.

New caps in, I have circled the six affected capacitors that were replaced.

Back together with the new caps, all working ok !

Hauppauge HD PVR wrong power supply.

A Hauppauge HD PVR that had been powered up using the wrong PSU.

With the correct PSU used again the unit would not work.

Unfortunately I didnt take any pics so I used one off the web to describe what I found, I hope the owner of this picture doesn't mind, it was from a blog post about modifying a unit by fitting a fan.

http://blog.arogan.com/2008/06/hauppauge-hd-pvr-model-1212.html

The picture does however raise an interesting question as it differs slightly, voltage regulator 4 was not present in this faulty unit ??

A good chunk of the circuit runs off 3.3v as you would expect, this seemed to be handled by the regulator #3 derived from the incoming 5v.

However some of the circuit runs off the 5v input straight frsom the wall adaptor, no circuit mount regulation or fuse.

Plug in the wrong supply and....

Pooof, at the very least IC's #1 and #2 were toast, #1 was long gone, nothing at all left of it. #2 was still present (a FMS6403 component video driver) but there was a hole in it.

I didnt spend any longer on the unit and told the owner the bad news, mostly as I could see no easy way to procure a FMS6403 and not knowing what chip #1 was. Chip #1 looked like it was a 3 terminal chip (Vin ground, Vout), there may have been some way to hack arround this.

Im sure some circuit protection would not be too expensive to design in. Hauppauge's idea of protection is a sticker on the back saying 'only use the supplied PSU' well, obvious but accidents do happen, searching the web showed that plugging the wrong PSU in was not an uncommon occurrence.