Q: PCB track spacing for 230V AC
Neil Mackie
What spacing should I use between the tracks to avoid arcing?
i.e. whats a safe figure in mm/100V for the track spacing.
Any other references I can consult?
Thanks,
Neil J Mackie
Dan Clingman
Les Nagy
regularly in air. I do not know the accepted practice for spacing of PC
traces, but there are some things to consider. First of all, 230VAC has
a peak to peak value of 650V, and that is assuming no spikes occur.
Moisture can also affect the arc over distance. For 230VAC, I would use
a minimum of 5mm spacing so that a wide safety margin exists and a
conformal coating can also reduce the chance of arc over.
--
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Les Nagy: President | mailto:ln...@netaccess.on.ca
|
Aldebaran Technological | Phone: (905) 388-1011
Services Inc. | Fax: (905) 388-9028
http://www.nas.net/~lnagy
John Woodgate
<N.Ma...@btinternet.com> writes
>I'm designing a pcb which I want to connect to 230V AC.
>What spacing should I use between the tracks to avoid arcing?
>i.e. whats a safe figure in mm/100V for the track spacing.
>Any other references I can consult?
You will find limits on track spacing in relevant safety standards, such
as BS EN60065, BS EN60950, BS EN60204, BS EN61010-1, BS EN60335-1. There
are usually special requirements for mains conductors, because of the
incidence of voltage spikes up to 2 kV (4 kV on industrial supplies).
--
Regards, John Woodgate, Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
OOO - Own Opinions Only. You can fool all of the people some of the time, but
you can't please some of the people any of the time.
Bugs
wrote:
>I'm designing a pcb which I want to connect to 230V AC.
>What spacing should I use between the tracks to avoid arcing?
We use (UK) 3mm as safety gap for mains to anything else.
George Smith
Return address invalid to prevent junk mail.
Email: bu...@mindless.com
Spehro Pefhany
> We use (UK) 3mm as safety gap for mains to anything else.
I thought 8mm creepage was required for Euro approvals?
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Spehro Pefhany "The Journey is the reward"
sp...@interlog.com
Fax:(905) 271-9838 (small micro system devt hw/sw + mfg)
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Richard Sulin
but that is for 120VAC, of course. Extending that logic would compel
me to reply "12mm"....
Rich
---------------------------
Neil Mackie wrote:
> I'm designing a pcb which I want to connect to 230V AC.
> What spacing should I use between the tracks to avoid arcing?
> i.e. whats a safe figure in mm/100V for the track spacing.
> Any other references I can consult?
>
> Thanks,
> Neil J Mackie
--
+-----------------------------------------
| Richard Sulin
|
| "under 3000 characters for Mr. Sharpe"
|
| Please remove the anti-spam stars in
| my e-mail address in order to reply.
|
| thus: richs <at> dcdu <dot> com
+-----------------------------------------
John Woodgate
<sp...@interlog.com> writes
>the renowned Bugs <nospa...@mindless.com> wrote:
>
>> We use (UK) 3mm as safety gap for mains to anything else.
>
>I thought 8mm creepage was required for Euro approvals?
>
address and decided to exercise restraint! This guy is living in a
fool's paradise if he really uses 3 mm for mains. I just hope I don't
have any of his products!
John Woodgate
<***richs***@***dcdu.com***> writes
>I was told by an experienced PCB designer that it's 6mm for UL (USA),
>but that is for 120VAC, of course. Extending that logic would compel
>me to reply "12mm"....
No, that's defintely OTT. 8 mm is OK everywhere, but smaller
clearance/creepage is allowed in some cases.
Tony Williams
> > We use (UK) 3mm as safety gap for mains to anything else.
> I thought 8mm creepage was required for Euro approvals?
I don't know what it is now, but seem to remember that
the old minimum acceptable creepage length for UK 240v
terminals was 1/4". On a pcb this would be the edge
to edge distance of the pads. But I hate mains on a
pcb anyway.
--
Tony Williams.
Bugs
>In article <72srk9$pfa$1...@news.interlog.com>, Spehro Pefhany
><sp...@interlog.com> writes
>>the renowned Bugs <nospa...@mindless.com> wrote:
>>
>>> We use (UK) 3mm as safety gap for mains to anything else.
>>
>>I thought 8mm creepage was required for Euro approvals?
>>
>In certain cases, yes, but not all. I noted the 'mindless' in the
>address and decided to exercise restraint! This guy is living in a
>fool's paradise if he really uses 3 mm for mains. I just hope I don't
>have any of his products!
<Sigh>...
But in the real world.
See Farnell catalogue Arcolectric 152-274 PCB mounting 240/115 volt selector
switch.
Approvals VDE, SEV, DEMKO ,NEMKO, SEMKO, KEMA, SETI, OVE, UTE, SAA, Ul and CSA.
Pitch between pins 6mm and 7mm.
Hole size 1.3mm
Annular ring... leaves 8mm???
Why do /so/ many engineers make me use these impossible products?
Euro?? Ah yes - the guys who decided a good frequency for a car door opener
would be right in the middle of a UK amateur (400 watts permitted) band.
--
bugs
Spehro Pefhany
> Pitch between pins 6mm and 7mm.
> Hole size 1.3mm
> Annular ring... leaves 8mm???
> Why do /so/ many engineers make me use these impossible products?
Easy enough (in quantity) punch the board with slots. There is probably a
divider molded into the plastic on the switch. Though 8mm is tough, it
isn't a bad idea for safety related situations, even if it is not strictly
required. For example consider a board, one trace connected to the 230V
mains, and the other to the finger of your wife etc. as she sits in the
bathtub or stands with wet feet on a metal plate. How much clearance would
*you* leave? (and if you are in the middle of a custody battle, your
answer doesn't count). ;-)
Bugs
>How much clearance would
>*you* leave? (and if you are in the middle of a custody battle, your
>answer doesn't count). ;-)
Drat! :(
But that's obviously medical apparatus which was not specified by the
originator. And yes we have done stuff for mines and bore-holes.
For the original poster - dredged up the following...
http://www.vutrax.co.uk/vbook1.htm
When high voltages are involved, the following UL derived rule provides the
minimum clearance between tracks with designated potentials between them at
normal temperature and pressures, or when coated:-
0.023" + (0.0002" * peak volts)
0.584mm + (0.005mm * peak volts)
For single phase UK mains (assuming Peak of 400V) we get 0.103" (2.58mm). Good
design practice in any case partitions PCB's into 'safe' and 'unsafe'
areas with wide separation, often coating the 'unsafe' side to reduce the hazard
to maintenance persons.
.............
and
From text book of PCB Design.
"Referring to IEC recommendation 348 Safety Requirements fo Electronic Measuring
Apparatus, the effective safety distance (creepage distance) for class I is 3mm,
(0.12 in) for a.c. voltages from 60V up to 250V. In the case of printed circuit
boards, since the position of the conductor cannot be changed, the safety
distance may be reduced to 2mm (0.79in) for all pads and conductors carrying the
mains voltage. National safety regulations can, however, set more severe
requirements than IEC. This is the case with UL safety regulations (UL,
Underwriters Laboratories Inc., USA) wher a safety distance of 1/16 in (1.6mm)
is required for a.c. voltages up to 125V and 3/32in (2.4mm) for a.c. voltages up
tp 250V."
<snip stuff about pcb tolerances and projections>
"Design Rule Number 3(B)
In the case of a.c. mains voltages up to 250V, the safety distance is ... 3mm
(0.118 in)."
--
bugs
Spehro Pefhany
> But that's obviously medical apparatus which was not specified by the
> originator. And yes we have done stuff for mines and bore-holes.
>
Sure. Of course if you make zillions of something, you can be sure that
someone will test the insulation in such a obviously (to a sentient being)
hazardous way.
<lots of good stuff snipped>
Ok, my addition from EN 60335-1
"For conductive patterns on printed circuit boards, except at their
edges, the values in the table between parts of different potential may be
reduced as long as the peak value of the voltage stress does not exceed:
- 150 V per mm with a minimum distance of 0.2mm, if protected against the
deposition of dirt
- 100 V per mm with a minimum distance of 0.5mm, if not protected against
the deposition of dirt
For peak voltages exceeding 50V, the reduced creepage distances only apply
if the proof tracking index (PTI) of the printed circuit board is greater
than 175 when measured in accordance with index N. "
So, for 400V peak, we have 4mm or 2.7mm, based on the above.
Richard Kilsby
<sp...@interlog.com> writes
big snip of good stuff.
>
>For peak voltages exceeding 50V, the reduced creepage distances only apply
>if the proof tracking index (PTI) of the printed circuit board is greater
>than 175 when measured in accordance with index N. "
>
I wondered when tracking indexes - comparative or proof would rear their
ugly heads. Does anybody actually have a definitive source for the
measurement or method of determination?
Circuit board manufacturers and actual laminate suppliers that I have
spoken to over the years consider it to be a black art. And not
repeatable.
A bit like the method of determining the alcohol content of beer by
whether or not the duty mans' leather breeches stick to a wooden stool
soaked in said beer whilst he finishes his pipe.
Also does the PTI or CTI apply when a solder mask is applied?
Does the mask get tested?
Rick Kilsby
John Woodgate
<nospa...@mindless.com> writes
><Sigh>...
>But in the real world.
>See Farnell catalogue Arcolectric 152-274 PCB mounting 240/115 volt selector
>switch.
>Approvals VDE, SEV, DEMKO ,NEMKO, SEMKO, KEMA, SETI, OVE, UTE, SAA, Ul and CSA.
>Hole size 1.3mm
>Annular ring... leaves 8mm???
>
>Why do /so/ many engineers make me use these impossible products?
I don't really see the point you are making, but if it is that this
component doesn't have 8 mm clerarance/creepage but carries numerous
approvals, then it is just that it is OK for those situations where 8 mm
is not required. Using it, even with all its approvals, in a situation
where 8 mm IS required, would not be acceptable.
John Woodgate
<nospa...@mindless.com> writes
>For the original poster - dredged up the following...
>
>http://www.vutrax.co.uk/vbook1.htm
>
>When high voltages are involved, the following UL derived rule provides the
>minimum clearance between tracks with designated potentials between them at
>normal temperature and pressures, or when coated:-
>
> 0.023" + (0.0002" * peak volts)
> 0.584mm + (0.005mm * peak volts)
>
>For single phase UK mains (assuming Peak of 400V) we get 0.103" (2.58mm).
No. Current safety standards take acount of voltage spikes that appear
on mains supplies. You need to consider 2 kV spikes on household 230 V
mains supplies, also that greater separation is required for Class II
('double-insulated') apparatus. In EN60065:1994, 3 mm is allowed for
Class I (protection by earthing) but 6 mm is required for Class II.
Other standards require larger values, especially where insulator
surfaces can be contaminated by dirt and/or oil.
>Good
>design practice in any case partitions PCB's into 'safe' and 'unsafe'
>areas with wide separation, often coating the 'unsafe' side to reduce the
>hazard
>to maintenance persons.
>.............
>
>and
>
>From text book of PCB Design.
>
>"Referring to IEC recommendation 348 Safety Requirements fo Electronic
>Measuring
>Apparatus,
That standard is obsolete. See IEC/EN61010 now.
> the effective safety distance (creepage distance) for class I is 3mm,
>(0.12 in) for a.c. voltages from 60V up to 250V. In the case of printed circuit
>boards, since the position of the conductor cannot be changed, the safety
>distance may be reduced to 2mm (0.79in) for all pads and conductors carrying
>the
>mains voltage.
EN60065:1994 has a similar provision (but excludes pads) and this has
been challenged as being unsatisfactory.
[snip]
The whole subject is under re-consideration by IEC TC28, and there is a
published Standard IEC60664-1 and a Technical Report IEC60664-5 in
preparation. These may lead to reduced separations being required in
future, but a much more searching analysis of the required clearance in
every case.
John Woodgate
<Ric...@four-c-elec.demon.co.uk> writes
>In article <731rae$6nc$1...@news.interlog.com>, Spehro Pefhany
><sp...@interlog.com> writes
>
>big snip of good stuff.
>>
>>For peak voltages exceeding 50V, the reduced creepage distances only apply
>>if the proof tracking index (PTI) of the printed circuit board is greater
>>than 175 when measured in accordance with index N. "
>>
>I wondered when tracking indexes - comparative or proof would rear their
>ugly heads. Does anybody actually have a definitive source for the
>measurement or method of determination?
>
>Circuit board manufacturers and actual laminate suppliers that I have
>spoken to over the years consider it to be a black art.
Well, 'black art' is often the test result!
> And not
>repeatable.
Not in polite company, anyway.
>
>A bit like the method of determining the alcohol content of beer by
>whether or not the duty mans' leather breeches stick to a wooden stool
>soaked in said beer whilst he finishes his pipe.
Worked very well for some hundreds of years!
>
>Also does the PTI or CTI apply when a solder mask is applied?
>Does the mask get tested?
>
See IEC60664-1 and, especially, -3, and IEC60664-5 when it is published.
Also, IEC60112, 60587 and 61302.