While we're on the subject of PD3, I have some 950-based questions regarding
optocouplers and creepage distances.
Suppose we have a power supply with a 230Vac supply and a user-accessible
secondary somewhere. If you put an optocoupler across the Pri-Sec isolation
boundary, it seems to me that the required creepage can easily exceed that
offered by standard DIP packages.
- The standard DIP gives you 7.4mm while the wide-body types can give you
up to 10mm.
- I haven't looked today, but I recall determining once that the packages of
opto's and other IC's are Material Group III
- We're discussing Pollution Degree 3
- Reinforced insulation is required in the above scenario
A little reverse engineering of the creepage table tells us that only 300V
RMS or DC working voltage is needed before the requirement exceeds the 10mm
provided on a wide-body package. Standard DIP's with 7.4mm creepage fail at
231Vrms!
Q1: Right so far?
Q2: Anyone have any differing info re the material group for opto packages?
Determination of the working voltage can be confusing. Suppose the primary
side of the opto is in a 5V logic circuit but that circuit's reference
ground is the same as the rest of the primary = the negative side of the
rectified line voltage bus (ie one diode away from the AC neutral). The
rectified line voltage bus operates at around 300Vdc (ignore ripple for
now). I think we need to consider the working voltage on the primary side
of the opto to be 300Vdc.
Q3: Is 300Vdc right?
Now suppose on the secondary side of the product we have a floating 5V
control circuit isolated from the primary by a housekeeping supply, but
referenced to the same (floating) secondary ground as the secondary power
circuits which include a 45Vdc bus. I think we need to consider the working
voltage on the primary side of the opto to be 45Vdc.
Q4: Is 45Vdc right?
The opto is connected between these two 5V control circuits: one referenced
to primary ground and one to a floating secondary ground.
Q5: What is the working voltage across the opto, and how do you arrive at
it?
Q6: If I'm right, the working voltage is 345Vdc and no opto I know of meets
the required creepage (interpolates to 11.2mm). Can I apply the easement of
par. 2.9.8 by using a conformal coating on the opto, therefore allowing the
Table 7 creepage of 2mm? In other words, are opto's considered to be
components conforming to 2.9.7?
Q7: If yes, do I need to do the tests of 2.9.7 to the opto (testing my 2nd
and 3rd sources too) or have the opto manufacturers been kind enough to do
this for me during the opto's approval process?
OK, enough already. I'll stop now and see what replies I get. Thanks in
advance for your help.
Regards,
Jim Eichner, P.Eng.
Group Leader, Engineering Services
Xantrex Technology Inc.
Mobile Markets
web: www.xantrex.com http://www.xantrex.com
Any opinions expressed are those of my invisible friend, who really exists.
Honest.
-Original Message-
From: Massey, Doug C. [mailto:masse...@ems-t.com]
Sent: Tuesday, April 03, 2001 2:10 PM
To: 'IEEE Forum'
Subject: RE: 950 Pollution degree detirmination
The UL interpretation is: Pollution Degree 2, based on the IP66 rating.
I think EN60950 3rd Ed., clause 2.10.1 is very open to interpretation - I
can make a great case that my power supply inside the enclosure is a
subassembly (what else could it be?). Also, that subassembly is sealed
against moisture or dust, as evidenced by the type test performed to a
harmonized standard. I do understand the intent of PD 1, though, as others
have responded in so many words, that's for things that are potted.
We're designing to PD 3 spacings - I've already played devil's advocate as
you did. I just wanted to hear a few opinions.
Doug
-Original Message-
From: John Juhasz [mailto:jjuh...@fiberoptions.com]
Sent: Monday, April 02, 2001 1:35 PM
To: Massey, Doug C.; 'IEEE Forum'
Subject: RE: 950 Pollution degree detirmination
Doug,
While I would tend to agree that this may be acceptable,
and I would too like to see some sort of 'official'
determination on this (I may have a similar
situation arise in the next couple of months),
I would like to play devil's
advocate for a moment to further this discussion . . .
As you quoted . . . EN60950 3rd Ed., clause 2.10.1 defines
Pollution Degree 1 for components and subassemblies which
are sealed so as to exclude dust and moisture (see 2.10.7).
It clearly states for components and subassemblies. Your
Pollution 1 argument requires the use of the IP66 enclosure.
But is the product/enclosure pairing still considered
a 'component' or subassembly'?
The way I see it is that the product without the enclosure
is the component or subassembly, and that installing
the product in the IP66 enclosure make the product a
'top-level'. If for some reason the gasketing on the enclosure
fails, or it is not closed properly, the product
would
