Even if our homebrew audio equipment is run from an official ClassII certified standalone PSU?
effectively you are building a mains isolated device. That puts you into a different regulation.
Are you designing a mains powered device or a low voltage device that is isolated from the mains?
"Common" is an even better term. Which made me me think you were making a pun here:
... having recently read this...
Can we find common ground about "common" and "ground"?
We are also not competent to do this for class I equipment, either!
Simply connecting PE to the casing doesn't negate the requirement for high voltage and insulation tests of a mains powered device, to name just one thing most of us won't be able to do at home...
If one were to apply rules and regulations strictly, we'd probably only be allowed to build non mains powered devices and power them from bought (certified) equipment (like slipstreem mentioned). Yeah right...
Last edited:
In my specific case, it tends to be mostly low voltage (<=20V DC) gear run from bought-in DI PSUs carrying the official DI logo, but as jitter says, we aren't really qualified for ClassI unless we have access to a certified earth bond/flash tester.
I and 50,000 Members do not agree with you.
A "Megger" would do a high voltage test.
A deliberate short could check that the protection actually functions.
But if you have a Megger, couldn't you also use that to test class II devices?
Using common sense, it shouldn't be too hard to construct a class II device... Just keep a good distance (> 6 mm) between metal casing and exposed mains carrying parts (e.g. pcb tracks), use a good strain relief on the (fixed) power cord (or alternatively use the 8-shaped mains entrance) and shrink tubing on exposed solder joints (switches, fuseholders). Oh, and the transformer should be class II.
I do it like this, but that's at my own risk, I'm not saying you should!
It isn't.
The hard thing is to prove that is complies with class II. One or several samples must be dropped from a defined height. Usually that means mechanical damage to or destruction of the sample. The sample still has to provide protection from touching dangerous voltage after that drop.
In mass production there is usually no problem in sacrificing one or a few samples. How many samples can you afford to drop when you DIY only a few amps or a single one?
The operative word being "prove".
You wouldn't be able to go the harmonised standards route to CE compliance, but unless you put it to business use, proving CE compliance is not necessary AFAIK.
For robustness, it shouldn't be difficult to examine some class II equipment and construct something unarguably more resilient.
To comply with IEC 60065 safety standards, you would also need to carry out vibration testing and humidity testing. to pretty exacting standards IIRC. That applies to both Class I and Class II equipment. The humidity test would be especially difficult with typical DIY facilities.
Even though I agree that building your DIY project as a class I appliance is a safer bet when trying to comply with regulations than building it as a class II appliance. But what some seem to suggest is that class II appliances need to be tested to the limit and class I would not. And that I don't agree with.
The company I work for not only makes pcbs but also mains powered appliances, as far as I know all class I. Each one of them is required (if I'm not mistaken: by law) to be tested for electrical safety. Either by us, or by our customer before going to the end user. This means at least a HiPot test. Some customers may also want additional tests like insulation resistance and ground bound tests (HiPot and IR are very similar, but HiPot is more sensitive to arcs and coronas (too tight spacing, damaged insulation) and IR is more sensitive to leakage caused by contaminants and humidity.
IMO a DIYer without the proper tools can't prove to comply with either class I or II. I agree with Omega_Void here. In both cases we're constructing something unproven. I'd rather construct it as class II.
I believe it's intrinsically safer to build something that prevents a fault ocurring (class II) than one that needs to rely on PE (class I) to "cure" it when it does. Most older buildings may not even have earthed sockets in "dry" areas in this country!
Plus added to that, I've had my share of groundloops related to PE (earthed PC one one side + TV connected to cable on the other = big groundloop and horribly buzzing amp). I have one class I appliance in my audio system (a DAC with symmetrical in- and outputs), and because of that I connect SPDIF from my AV-sources (PE on DAC + cable-TV = another groundloop) to the DAC through Toslink only.
The company I work for not only makes pcbs but also mains powered appliances, as far as I know all class I. Each one of them is required (if I'm not mistaken: by law) to be tested for electrical safety. Either by us, or by our customer before going to the end user. This means at least a HiPot test. Some customers may also want additional tests like insulation resistance and ground bound tests (HiPot and IR are very similar, but HiPot is more sensitive to arcs and coronas (too tight spacing, damaged insulation) and IR is more sensitive to leakage caused by contaminants and humidity.
IMO a DIYer without the proper tools can't prove to comply with either class I or II. I agree with Omega_Void here. In both cases we're constructing something unproven. I'd rather construct it as class II.
I believe it's intrinsically safer to build something that prevents a fault ocurring (class II) than one that needs to rely on PE (class I) to "cure" it when it does. Most older buildings may not even have earthed sockets in "dry" areas in this country!
Plus added to that, I've had my share of groundloops related to PE (earthed PC one one side + TV connected to cable on the other = big groundloop and horribly buzzing amp). I have one class I appliance in my audio system (a DAC with symmetrical in- and outputs), and because of that I connect SPDIF from my AV-sources (PE on DAC + cable-TV = another groundloop) to the DAC through Toslink only.
Last edited:
With class II you only have the choice to test to the limit or not at all. With class I you can do some tests that may not comply with regulations, but give a satisfactory indication that the device is safe. E. g. measuring the earthing resistance with a DMM will not comply with regulations, but gives a good clue about the quality of the PE-related connections.
You're welcome.
Since you're amp is in a plastic case (effectively EMI unshielded), I wonder if it wouldn't be better to use coax in the line level part of the signal, shield connected to signal-GND. I.e. between input, pots and pcb.
OK, since the problem has been fixed now, on with the discussion...
What do you suggest to use instead of "common"? "Earth" and "ground" are also confusing, so those are out too...
Congratulations, and thanks for telling us exactly what the problem was, as we now know one more thing to look for when a similar problem comes up.
"Circuit signal/circuit power return path" ? That phrase isn't used in the electrical power industry, is it???
With the caveat about "common" being used as "hot" in electrical power transmission, I can see the concern of using it in electronics for "circuit ground," but this (using a, er, "common" word for a specific technical meaning) happens a lot, and it's not surprising that such a word gets FURTHER overloaded by having several specific technical meanings in related fields.
The "circuit ground equivalent" in electrical power distribution would be neutral, but neutral is obviously a bad name for a "circuit ground" as one SHOULD NOT connect the two (except of course in the special case of devices directly powered off the power line).
Also, for clarity in the direction this thread has taken, I wasn't sure of the difference between "Class I" and "Class II" so here I reference "The Repository Of All Human Knowledge" on the topic:
Appliance classes - Wikipedia, the free encyclopedia
I notice the Wikipedia article references international standards from IEC. Perhaps that entity should be the one to come up with a name for "electronic circuit
This is a perennial "known problem" with Cable TV, as the cable is grounded at the utility pole instead of to the ground of the building it's feeding (and even there, ground loop hum is easily possible). Fortunately, the signal is all RF and a simple, small RF isolation transformer to remove the galvanic connection fixes the problem. Those things are cheap enough that every cable TV installation should have one of those where it enters the building, especially with how many cable TV cables are connected to "home theater" audio systems thesedays.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.