forum safety rule - question

Status
Not open for further replies.
Jump to Latest
Bigun

Bigun

Member
Joined 2009
diyaudio.com/community/threads/tgm8-an-amplifier-based-on-rod-elliot-p3a.245619/
Paid Member
#1
I note that we are not allowed, for good reason, to post designs for circuits that are directly connected to the AC mains.

However, we do include the design and construction of linear power supplies which are directly connected to the mains. The basis for safety is the isolation provided by the power transformer, built to one of the many international safety standards.

I would like to understand if a similar level of safety and mains isolation is possible using a Y rated safety capacitor and if such an approach would be approved for discussion on this forum.

These capacitors are designed for use in series with the live mains and local ground, in applications where a failure (short) would create a risk of shock to the user. They are built to a number of international safety standards. For example, Y1 class are rated to 500 Vac and Y2 class for 150 Vac to 300Vac and it is best to use Film types.

The attraction might be for low power h.t. supplies for vacuum tubes.

Thoughts ?
 
#5
Something else to be aware of is the required isolation distances between live mains and the rest of the circuit. Using a transformer almost guarantees this without too much thinking as the terminals are spaced apart. Using a cap means that mains and the rest can share a PCB, so the designer needs to be aware of spacing requirements. I seem to recall that Europe requires a minimum of 4mm gap - although often ignored by China Export items.
 
#6
SELV
Spacing classification for mains and other high voltages is a fun area, there are so many specs such as IPC-2221, IEC-60950, EN-50124-1/2, IEC-61010 (UL versions have a UL in the name)
Not only does clearance have to be considered but also creepage distance which can be 2X the clearance.
This site is a good start...
WWW.CREEPAGE.COM Welcome!

But like EMC considerations I have had a far from enthusiastic response to any safety or EMC comments, often with comments such as "it is DIY so doesn't matter" or similar etc.
I would place all DIY kit for domestic use in Pollution class 2 for the purposes of calculating any spacing requirements, and all AC voltages are calculated from the peek not RMS.
Pollution Classes....
Pollution Degree Rating for Electrical Equipment - National Instruments

The fun is trying to plow your way through the specs, the main thing to remember that creepage and clearance distances quoted are for "THROUGH AIR" so the same rules do not apply to PCB inner layers or any insulated conductors. Just done a BIG SMPS/Invertor (to power all the devices in a railway carriage such as laptops etc.) where for reasons of safety all HV copper was on 6oz inner layers, se we could get a 2.5K isolation barrier for over a minute.
As most DIY designs are 2 layer for reasons of cost then the rules and spacing's should always be adhered to.
 
#7
Allowing discussion/use of Y type capacitors potentially allows another design choice. I don't think it will necessarily save money but can save weight, space and magnetic flux leakage.

Discussion provides better awareness of safety aspects - afterall people on the forum sometimes restore old radios (transformerless), or use soft-start modules from eBay which are based on capacitor isolation to derive relay coil energy, or want to improve mains filtering (a common application of X and Y safety caps)

Lead / creepage rules apply also to pcb mount transformers - a good topic for DIYers to learn about.
 
#8
Hi,

Talking nonsense is not a good guide to safety.

Y capacitors are not used in series with anything.
X capacitors go across live and neutral.
Y capacitors go across live to chassis and neutral to chassis.

I fail to see how anything in parallel gives any safety
features. No transformer is still lethal. Using the caps
only means your protected from mains lightning strikes.

If the intention is to suggest Y caps used in series with live
allows safe low current directly connected mains it should
be made clear, no way will a typical Y cap in series allow
a transformerless radio to draw enough current.

rgds, sreten.
 
#9
This has the potential to lead to an unsafe condition. A single Y cap of 4700pF between the hot side of the line, and a grounded user, should not deliver a lethal shock under normal circumstances. It will not deliver enough current to power anything useful either. It is possible to see how the current will increase if the line voltage is badly distorted, but again, probably not lethal.

I'm wondering if this will lead to someone thinking that they are protected if they connect 100 Y caps in parallel. After all they are not directly connected to the power line?

We see power supplies that are connected directly to the line on this forum all the time. There used to be a common schematic where one SECONDARY was connected to the line, while another SECONDARY was connected to the device circuitry. This relies on nothing more than wire insulation for mains isolation. I have repeatedly stated that this is not safe.

Some of us have begun to experiment with line powered SMPS designs. For now these are just black boxes that are not discussed. Sometime we should discuss the proper way to handle this subject on this forum, since many users are clearly not qualified to go there, but some are.

Line powered devices are all around us, including most modern TV sets. How do we discuss this subject in the future?
 
#12
sreten said:
Talking nonsense
....Y capacitors are not used in series with anything.
Click to expand...

Let me give some context: The Y capacitor is usually placed physically 'across' line and ground conductors but electrically I also view it as a series circuit whereby current flows from line though the cap to ground and back to the generator - a current loop.

It's my perception from reading around that your posts are sometimes unnecessarily challenging and negative.
 
#14
marce said:
A prime example of what I would consider beyond DIY and unsafe in the extreme...
http://www.diyaudio.com/forums/power-supplies/125269-50-v-smps-10.html#post4335298
Rectifried direct mains........

Whenever you have mains on a board or other high voltages there should be a clear defined separation between the HV side and the low voltage side, through the centre of the transformer.... as large as possible.
Click to expand...
yer a funny bird marce, last week you champion using 'offline' SMPS and then today you damn them in another thread. 😉IMO some DIYers are qualified to build "extreme" things and others are clearly not, that's why they have a report button, it lets an 'expert' make the call. But I agree some beginners have no sense building >300W stuff. It takes a few years experience.
I hope the experts will go quiet when a noob says a built this huge-n- dangerous thing and it doesn't work now! but global censorship of images in the PS forum would be bad thing too.
 
Last edited:
#15
The basic concept is this (my own opinion, NOT speaking as an agent for the forum): if you're endangering yourself with your own stupidity, well, that's evolution in action. If what you're doing can reasonably endanger others (e.g., knobs or patch cords or whatever) because of your stupidity, that's a different matter. A dead kid who was unfortunate enough to touch an innocent-looking volume control knob is far more horrible to me than some dipstick who builds a 3000 volt supply for his first project and sends himself across the Rainbow Bridge.
 
#17
I've built some simple things like strobe lights running directly off-line, using the reactance of a capacitor to drop the voltage, rather than a burning hot resistor. It's efficient, but the cap is overly large and any line transients just blow right through. It's not usually a good way to go, and I ultimately changed over to a 24 VAC wall wart and cap coupled that. Transformers are your friends!
 
#18
Bigun said:
Let me give some context: The Y capacitor is usually placed physically 'across' line and ground conductors but electrically I also view it as a series circuit whereby current flows from line though the cap to ground and back to the generator - a current loop.

It's my perception from reading around that your posts are sometimes unnecessarily challenging and negative.
Click to expand...

well yes, but only to drain rfi to ground and not a part of the normal working of the amplifier....

i see this done on a lot of atx psu's.....that is where i get my x and y caps from....😀

on my side of the pond, you can get heaps upon heaps of discarded atx psu's for almost nothing...
 
#19
Bigun said:
It's my perception from reading around that your posts are sometimes unnecessarily challenging and negative.
Click to expand...


Hi,

Yes they are challenging because I want people to be able work
it out for themselves, they are negative because assumptions
are often simply wrong and need to be looked at more severely.

Here the issue is what Y capacitors actually do beyond EMI.
In no way are they in series in any sensible practical sense.
Its their self healing breakdown as a parallel device that matters.
(And that is basically lightning strike protection, nothing else.)

In no way do they guard against the perils of a direct mains supply,
which is relatively unlimited fault current, they have no effect.

rgds, sreten.
 
Last edited:
Status
Not open for further replies.
Top Bottom