DIY IEC Protection class II units? How to deal with tube voltages in a device without PE?

[rock soderstrom]

What are you going to test that relates to eg. 60065 or other concerns? Is there a plan, or are you just going to ask some questions along the way ?

There is actually a "plan" that I would like to outline tomorrow. This is of course not a professional project, but a DIY project which I will publish here again in a dedicated thread. (if everything works out).

But I'm really interested in building it safely and discussing the pitfalls and regularizations on the basis of this project. This also includes the points you mentioned.

For me it is easier and more interesting to learn with a real project.

A little patience please, it's almost 3.30am here now, it's time for sleep mode.

 

[warpie]

Safety is half of the story and perhaps, the "easy" stuff.

EMC and EMI is a bloody complex and very pricey rabbit hole.

 

[rock soderstrom]

Safety is half of the story and perhaps, the "easy" stuff.

EMC and EMI is a bloody complex and very pricey rabbit hole.

I think you're right, that will certainly become an topic.

Before I continue with the safety discussion I wanted to check the PSU board more closely. Does it work adequately and can it deliver the promised values? What is the thermal situation under load?

I have built up the almost final circuit for testing this. Ignore the shifted component print, was my first attempt.

Everything looks good so far. I'm feeding the whole thing with a 12V/ 2000mA wall wart switcher.

At the input of the PSU there is therefore 11.9V DC, as a load my circuit has 600mA on the heater and approx. 10mA for the B+ circuit.

This results is a heater voltage of 6.6V and a B+ of up to 297.4V! Thus the PSU successfully fulfills its target specifications of 150-280V/ 15mA and 6.3V/ 1500mA so far.


Thermally, the whole thing is inconspicuous so far.

Let's continue, this is how the basic layout should be later on.

The idea is to build the fx device compactly

 

[mihi_fuchs]

I do not know if this topic is still relevant to you, but just stumbled over it. I found the explanaitions in the IEC EN 60335 quite alright regarding the topic (general standard on the electronic safety of household devices). They describe the testing you have to comply with and the options (eg. plastic housing and requirements, or metallic housing + inner isolation). These designs are much more common in usual household designs than in proaudio. Therefore the IEC EN 62368 (replacement of the IEC EN 60950) is very brief with regards to that, and I could not get a good picture out of the things mentioned there regarding safety class II. But others may be smarter.
In case you do self etching you can also find the required clearance and creepage distances in there. With tube devices, although I think clear to everybody, the tube counts as electronic part with voltage too, as the glass does not comply with the requirements towards isolation. So bulbs in the open are a topic of course.

Hope the 60335 may be a good hint. Otherwise it is at least worth to mention in the context.

Best regards,
looks really cool - how does it sound?

Michael

 

[rock soderstrom]

I do not know if this topic is still relevant to you,

Hi Michael, yes still relevant!

Hope the 60335 may be a good hint. Otherwise it is at least worth to mention in the context.

Very interesting, thanks a lot! I'll have a look at that after my vacation

looks really cool - how does it sound?

I worked on the housing last week, the next step will be the front panel. The reverb sounds good, it's definitely the easiest and cheapest way to build a tube spring reverb. (--> Schwabenhall ). I plan to publish the project here.

Cheers

during veneering

before danish oil

 

[rock soderstrom]

I still had some trouble with high frequency noise, of course I first thought that the switcher PSU is guilty, but it was the pentode that started to oscillate on a whim. Additional gridstopper was necessary...

 

[Tubetec]

I think with a sealed wooden box you could very easily end up where the thermal settling time is measured in hours or days and everything inside will be subject to needlessly high temperatures ,
Remember the sound of your springs will also change with temp as they expand and contract .

A metal tightly fitting box can at least loose some heat via conduction to the outside world ,
but wood is an excellent thermal insulator .

Who knows, a reverb tank cooking at 80C or more might sound like the sweat is dripping off the walls , but I dont think the electronics will live very long under those conditions .

heiß-halle might be a more appropriate name for it ,

 

[ruffrecords]

Have you worked out the total power consumed? There are only two tubes. I would be surprised if the total consumption was more than 20W. Even a sealed box would not get very warm inside. You could always add a small ventilation grill top and bottom, each with a simple labyrinth to stop things getting in, to encourage some airflow.

Cheers

Ian

 

[rock soderstrom]

I think with a sealed wooden box you could very easily end up where the thermal settling time is measured in hours or days and everything inside will be subject to needlessly high temperatures ,
Remember the sound of your springs will also change with temp as they expand and contract .

A metal tightly fitting box can at least loose some heat via conduction to the outside world ,
but wood is an excellent thermal insulator .

Who knows, a reverb tank cooking at 80C or more might sound like the sweat is dripping off the walls , but I dont think the electronics will live very long under those conditions .

heiß-halle might be a more appropriate name for it ,

What are you talking about?

Of course the cabinet has convection ventilation. You don't even know the plan for this reverb, but that doesn't stop you from telling stupid things, again. Three aluminum plates (front, back and bottom) are milled into the wooden housing with corresponding ventilation. Where do you think I should feed the power into the device? Where will the fuse holder and the on/off switch be attached? Where will the potentiometers and input and output sockets be placed? What is this about, are you on the booze again?

BTW, this unit doesn't even get lukewarm.


I think this is about something else, isn't it? Please embarrass yourself somewhere else and don't derail this thread. Your conspiracy theories and stupid talk are really hard to take, I will put you on ignore list from now on, if you have anything else to say to me, send me a PM

 

[Tubetec]

As usual Rock takes a little constructive criticism terribly personally and responds with a needless personal attack .

How am I to know what your final constructional details are , the unit isnt built yet .

My advice to anyone building tube gear is never to leave air flow and cooling as an after thought .
Tube mics are an obvious case where you might not have any airflow but its still worth considdering how the tubes heat is conducted away into the mic body and frame and away from other components .

 

[JohnRoberts]

Come on kids, play nice....

JR

 

[ruffrecords]

I just came across some additional information from a poster on the Kicad User group an a related but different issue. He says:

I don't agree with that, and I am involved in safety standards. Even if signal ground jumps to a high voltage due to a fault, the metal case is connected to chassis ground, which (in Class 1 equipment) is connected to the building safety ground. There is no electric shock hazard. If something has a metal case, it should be either Class 1 or Class 3 (powered by extra-low voltage). Class 2 ('double insulated') boxes should be non-conducting, even if they require an internal conductive coating or an internal foil shield for EMC reasons.

This seems to say that we are talking about a class 3 product rather than class 2 because class 2 has to have a non-conducting enclosure.

Cheers

Ian