Intermittent Potterton? Flashing red LED? PCB Fault finding

[geoff@cetltd]

Lea,

Suprima pcbs all eventually discolour due to the zener doides (please note the correct spelling) getting hot. On earlier pcbs they were not raised off the pcb. It is rare for these to fail

The relays suffer from what is called relay hammer which cracks the joints on the coil pins

The molex connectors fail because of poor quality flow soldering

As a design engineer, you might be suffering from the false impression that pcbs are built to a quality standard, err ... join the real world

I might suggest that you don't actually try and trace some of the circuitry through, it could be injurous to your mental health

I hope the pcb you bought off ebay caame with atlleast a 12 month warranty !

 

[Agile]

"""due to the zener doides (please note the correct spelling)"""

Well, if that is the correct spelling then "diodes" must be even more correct !

Quite apart from the mechanical shock when relays operate there is also the possibility of current carrying contacts overheating when they get dirty and desoldering themselves!

Tony

 

[geoff@cetltd]

"""due to the zener doides (please note the correct spelling)"""

Very very oops !

It's the beer in the keyboard - honest

but ... an electronics design engineer should be able to sprll zener correctly, even if another one always makes typos

Quite apart from the mechanical shock when relays operate there is also the possibility of current carrying contacts overheating when they get dirty and desoldering themselves!

Really - no - rarer than a very rare thing

it would melt the plastic of the relay casing first

Have you never come across a Stelrad 7A pcb ?

 

[Agile]

It may be rare to you but I have encountered it several times.

It normally affects the fan relay when there is APS fault or flame detection fault and the fan is cycling on/off.

I have to throw the PCB away and fit a new one!

Tony

jakata wed or fri

 

[geoff@cetltd]

Actually I have to take that back, don't I

Profiles and Netaheats suffer from this, having thought about it

email on its way regarding the Jakarta

 

[Gasguru]

Any good reason why the boiler industry hasn't progressed to snubberless triacs - are they too reliable or is there something in the gas appliance directive that insists on relays

 

[geoff@cetltd]

Inherent reliability is the last thing boiler manufacturers want.

there are numerous examples of "why the hell did they do it that way"

I came across a good example yesterday - on a Halstead Balmoral pcb, they used a 100megohm resistor in the flame sense circuit the sort of value that would be affected by a change in humidity

 

[Agile]

There is actually a reason NOT to use triacs.

They are most likely to fail short circuit which is the last thing the makers want on a gas valve.

A relay usually fails to the off position ( unless the contacts have welded ). Since the gas valve currents are usually less than 0.1A the chance of welded contacts is therefore small on an 8A relay.

Triacs are very sensitive to over voltage spikes whereas relays dont even see them!

Tony

 

[Gasguru]

Ideal Response has 7 relays from memory.

Many years ago as a student I worked at International Rectifier testing hockey puck triacs. They were pulse tested at 10,000 Amps!.

I sat about 2 meters from the transformer, the cables would jump off the bench.

These were primarily used for lift shaft motor control. Clearly they were thought sufficiently reliable enough to do the job.

 

[LeaUK]

No I didn't buy some rubbish from eBay, I reflowed the board myself, that means I control the level of quality.

An interesting point about solder joints and relay chatter but in my 20 years of designing (with relays) have never experienced it, but I'll bear it in mind. Though most my boards are double sided including PTH that naturally provide more mechanical support and are carefully designed for the mass Industrial and Medical markets.

Whilst I appreciate the significant cost requirements of consumer equipment, any Design Eng. that thinks about QCD and DFM MUST deal with these issues during development and they should not intentionally design circuits that discolour the PCB due to excessive heat! This is simply poor practice. Period.

Open up the cheapest DVD player's or PC SMPS and look and learn how DFM and QCD play major roles in modern design practice - I do regularly.

Anyway, I passed through this forum in the hope of some guidance from the more knowledgeable (and many thanks), to help and share my findings for the good of others and not really to discuss the merits of PCB design, so I'll leave you 'gurus' discussing.

BTW - raden, before criticising someone's spelling, do please read your own first! - I can recommend ieSpell


Cheers
Lea

 

[kier]

please dont tell subject author about puma pcb's that go on fire.

 

[geoff@cetltd]

"BTW - raden, before criticising someone's spelling, do please read your own first! - I can recommend ieSpell "

terrible, isn't it ... recommend WTF you like

my spelling's good, it's just the typing that isn't

I blame sticky beer in the keyboard, me

 

[meldrew's_mate]

please dont tell subject author about puma pcb's that go on fire.

...Or Suprima (Honeywell) gas valve solenoid coils which short to earth intermittantly and blow the supply fuse.

Anyhow, a facinating discussion, devoid of bitching (apart from the spilling orrers), which I found both educational and illuminating - thanks to all contributors.

 

[geoff@cetltd]

"please dont tell subject author about puma pcb's that go on fire"

I think I actually posted a photo some time ago

 

[ollski]

please dont tell subject author about puma pcb's that go on fire.

Never had that, but have had a fan sieze on an ideal elan which kept trying to spin it, melted the relay and set the board alight.