neon flasher problem

[[email protected]]

Hi guys,

I am facing a severe design problem. I have designed a simple flasher
circuit to flash a neon sign. The sign board has 5 neon transformers
driving the tubes.Each transformer is rated 7500-0-7500,30mA consuming
450VA from 230V AC mains. The flashing pulses are generated by a small
PIC. In my first design I used a 30A open frame relay to switch power
to the 5 transformers. My ignorance about relays led to contact failure
due to severe arcing in just 3 days ( sign board works for only 6 hours
a day). Then I tried zero crossing firing of the relays to minimise
arcing. It seemed to be ok with 2 transformers but at full load of 5
transformers it again started to arc.Fed up with relays I moved on to
triac...(I had only worked on maximum 200W resistive loads with
triacs). I used a BTA41 triac which is supposed to withstand 40A
current. An MOC3083 was used to provide zero-crossing trigger to the
BTA41. A snubber was also connected ( 0.01uF,630V and 100ohms) across
the traic.A suitable heat sink was also provided. Now bingo...the
flasher worked perfectly....but...not for long...It lasted 6
days...Can some body tell me what could have gone wrong??..or ....did
I miss any crucial design aspect?...Someone please help me!!

Thanks in advance,

Rasi

 

[[email protected]]

Hi guys,

I am facing a severe design problem. I have designed a simple flasher
circuit to flash a neon sign. The sign board has 5 neon transformers
driving the tubes.Each transformer is rated 7500-0-7500,30mA consuming
450VA from 230V AC mains. The flashing pulses are generated by a small
PIC. In my first design I used a 30A open frame relay to switch power
to the 5 transformers. My ignorance about relays led to contact failure
due to severe arcing in just 3 days ( sign board works for only 6 hours
a day). Then I tried zero crossing firing of the relays to minimise
arcing. It seemed to be ok with 2 transformers but at full load of 5
transformers it again started to arc.Fed up with relays I moved on to
triac...(I had only worked on maximum 200W resistive loads with
triacs). I used a BTA41 triac which is supposed to withstand 40A
current. An MOC3083 was used to provide zero-crossing trigger to the
BTA41. A snubber was also connected ( 0.01uF,630V and 100ohms) across
the traic.A suitable heat sink was also provided. Now bingo...the
flasher worked perfectly....but...not for long...It lasted 6
days...Can some body tell me what could have gone wrong??..or ....did
I miss any crucial design aspect?...Someone please help me!!

Thanks in advance,

Rasi

If you use relays you need snubbers accross the contacts and a suitable
vdr. You need much the same accross your triac, triacs dont allways
turn off verywell with inductive loads antiparallel scrs might be
needed, to reduce inrush you need to fire the triac at 90deg not 0deg
and keep it on with either dc or a stream of pulses.

 

[John B]

Hi guys,

I am facing a severe design problem. I have designed a simple flasher
circuit to flash a neon sign. The sign board has 5 neon transformers
driving the tubes.Each transformer is rated 7500-0-7500,30mA consuming
450VA from 230V AC mains. The flashing pulses are generated by a small
PIC. In my first design I used a 30A open frame relay to switch power
to the 5 transformers. My ignorance about relays led to contact
failure due to severe arcing in just 3 days ( sign board works for
only 6 hours a day). Then I tried zero crossing firing of the relays
to minimise arcing. It seemed to be ok with 2 transformers but at
full load of 5 transformers it again started to arc.Fed up with
relays I moved on to triac...(I had only worked on maximum 200W
resistive loads with triacs). I used a BTA41 triac which is supposed
to withstand 40A current. An MOC3083 was used to provide
zero-crossing trigger to the BTA41. A snubber was also connected (
0.01uF,630V and 100ohms) across the traic.A suitable heat sink was
also provided. Now bingo...the flasher worked perfectly....but...not
for long...It lasted 6 days...Can some body tell me what could have
gone wrong??..or ....did I miss any crucial design aspect?...Someone
please help me!!

Thanks in advance,

Rasi

Try SCR based solid state relays with zero-crossing detection, such as
Crydom EZ240D12S (one for each transformer would be best). Should be OK
without snubbers but for all the cost you might want to put a Semikron
SKRC440 across each relay.

 

[Wolfgang Mahringer]

Hi Rasi,

...An MOC3083 was used to provide zero-crossing trigger to the
BTA41. A snubber was also connected ( 0.01uF,630V and 100ohms) across

0.01 uF seems much too less for a 6x450VA snubber.
Try using 0.47 uF (may be even more) and 100 ohms.
Triacs and relay contacts will love it

HTH
Wolfgang

 

[Mark]

Rasi,

you may have an interesting problem regarding the transformers

when you suddenly turn off an iron core transformer, it retain the flux
level it was at at turn off in the core. The next time you turn it on
,the flux level continues from where it left off....

If by chance, the xformer is turned off at the peak of the AC wave, it
will retain a large flux level. If by chance it is turned on at the
zero crossing , the flux level will try t oincrease from the high
starting point and can possible go into saturation. If this happens,
the inductance goes way down and thew current way up. This lasts for a
cycle or two and is very damaging to the contacts.

Solutions? Some kind of soft turn off and soft turn on.. maybe a low
wattage incandescent bulb wired ACROSS each relay contact ( the
incandescent light will be on when the neon is off.) This will allow
the core to turn off softly and become demagnizied and will allow some
low level of current to flow. Also maybe a slow start PTC device in
series with each transformer to limit the inrush turn on current.


good luck

Mark

 

[ehsjr]

Hi guys,

I am facing a severe design problem. I have designed a simple flasher
circuit to flash a neon sign. The sign board has 5 neon transformers
driving the tubes.Each transformer is rated 7500-0-7500,30mA consuming
450VA from 230V AC mains. The flashing pulses are generated by a small
PIC. In my first design I used a 30A open frame relay to switch power
to the 5 transformers. My ignorance about relays led to contact failure
due to severe arcing in just 3 days ( sign board works for only 6 hours
a day). Then I tried zero crossing firing of the relays to minimise
arcing. It seemed to be ok with 2 transformers but at full load of 5
transformers it again started to arc.Fed up with relays I moved on to
triac...(I had only worked on maximum 200W resistive loads with
triacs). I used a BTA41 triac which is supposed to withstand 40A
current. An MOC3083 was used to provide zero-crossing trigger to the
BTA41. A snubber was also connected ( 0.01uF,630V and 100ohms) across
the traic.A suitable heat sink was also provided. Now bingo...the
flasher worked perfectly....but...not for long...It lasted 6
days...Can some body tell me what could have gone wrong??..or ....did
I miss any crucial design aspect?...Someone please help me!!

Thanks in advance,

Rasi

You could try one MOC3083/BTA16 triac/snubber at each transformer.
You don't have one inductance created peak to charge the cap,
you have one from each transformer.

Ed