|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
7th Aug 2016, 12:28 am | #61 |
Nonode
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,944
|
Re: The deeper mysteries of 'ground'?
Returning to the original post, attached are three papers that I downloaded sometime back, and which, in treating their primary subjects, do step somewhat into the “deeper mysteries” in a contextual way. This is a topic that I think benefits from studying multiple sources and looking at from multiple viewpoints.
Cheers, |
7th Aug 2016, 9:47 am | #62 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: The deeper mysteries of 'ground'?
I think it's worth emphasizing, as we have a mainly UK-based readership, that the three papers linked above refer to USA practice and some of the terminology, regulations, materials and configurations relating to electrical power and safety are different to UK equivalents. Although the physics is the same on both sides of the Atlantic and the two sets of regulations achieve similar aims, their methods differ in detail so what is permissible in one may be prohibited in the other. Electrical installation work in the UK should meet the current version of BS7671 and may fall under building regulations, which are statutory unlike BS7671 itself.
|
8th Aug 2016, 2:13 pm | #63 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
Quote:
Cheers!
__________________
Al |
|
12th Aug 2016, 2:46 pm | #64 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
Quote:
However in its present (original) configuration it also has 180VDC unregulated, in which the chassis is ground. We can clearly see the case of the original metal stud diode returned to chassis ground, as is the case of the big electrolytic. There is no safety earth. All parts of the case were clearly designed to be touched as the front panel is contiguous with the rest of the chassis. The 'safety' of this arrangement entirely relies on the fact the IR-type metal stud diode, popular in that era and a little beyond, has its cathode bolted to the chassis (conventional, I know, except that such a chassis doesn't have to be exposed.) This is clearly wired direct to mains neutral. However, if this component goes open-circuit, things are immediately rather dangerous: all other voltages are floating except mains, so in the event that there is also fault in the neutral circuit, a live wire to the chassis will not blow a fuse. Ouch.
__________________
Al |
|
12th Aug 2016, 2:50 pm | #65 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: The deeper mysteries of 'ground'?
No 'benchtop' equipment has any connection between the exposed chassis and either line or neutral. It has been forbidden since before there were electronics! It would also make the unit more or less unusable. If you think that is how it is arranged, there is almost certainly a mistake somewhere.
|
12th Aug 2016, 2:57 pm | #66 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
Quote:
You can clearly see a huge bunch of wires (white) terminating together in the upper part of the connector shown. These are also returned to chassis - take a look at the small bare wire bolted to the chassis next to the connector. And this is also the same ground as the -ve of the main reservoir and smoothing cap and the return for the big metal stud diode - this is very clearly bolted to the chassis, as shown in my last post. So yes, it does appear to be arranged this way. The front panel shown is also contiguous with the rest of the chassis. I'm not familiar with the type of push-fit connector whose reverse side is shown - it has multiple in-line brass connectors on the side not shown and is male.
__________________
Al Last edited by Al (astral highway); 12th Aug 2016 at 3:05 pm. |
|
12th Aug 2016, 3:15 pm | #67 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: The deeper mysteries of 'ground'?
I don't doubt lots of stuff is connected to the chassis, but which of those is mains neutral?
|
12th Aug 2016, 3:34 pm | #68 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
It's a grotty old mess of black tarry deposits on most of the wires, and I haven't traced the circuit. I'd have to clean it up and run a few more checks.
I'm sure you're right, I know you're expert in this area. If I'm mistaken, as I seem to be, the only plausible explanation is if the -ve rail of the 180VDC supply is returned to chassis, but in fact is also returned to safety earth, as are all those other white wires. This would make sense in the conventional and expected arrangement where the 180VDC is supplied upstream by a centre-tapped winding from the main isolating transformer - duh!! I'll have to do some deep cleaning around this, but I think we're back on track. Thanks, Lucien!
__________________
Al Last edited by Al (astral highway); 12th Aug 2016 at 3:54 pm. |
12th Aug 2016, 11:19 pm | #69 |
Heptode
Join Date: May 2015
Location: Bradford, West Yorkshire, UK.
Posts: 512
|
Re: The deeper mysteries of 'ground'?
At least it appears to have a 100K bleeder resistor, no such things in modern SMPSU's.
__________________
worried about the electrons entering the circuit and the smoke leaving Andrew |
23rd Aug 2016, 8:34 pm | #70 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
Any views, please, on using some thick brass sheet as a ground plane in a high voltage power supply? I have seen brass stanchions used as conductors in a pulse application - an induction heating circuit, and I have some wonderful vintage brass sheet lying around. I have also seen brass used in a 1940s signal generator.
The main connections would be made with lugs bolted with M3 bolts.
__________________
Al |
23rd Aug 2016, 8:42 pm | #71 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 14,007
|
Re: The deeper mysteries of 'ground'?
You say high voltage, but what sort of current?
Using 'thick' materials is only really relevant if you're going to be having significant circulating currents at DCV or low frequencies (remember the 'skin effect' which kicks in once the Megahertz begin to rise). |
23rd Aug 2016, 11:20 pm | #72 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
It's around 500mA average, big old peaks, though
__________________
Al |
25th Aug 2016, 8:07 am | #73 |
Hexode
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 407
|
Re: The deeper mysteries of 'ground'?
That's nothing. Save your brass for something worthwhile.
|
26th Aug 2016, 12:57 pm | #74 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
My query was really about whether it's a sufficiently good conductor ...if it is, I'm using it in this project . It needs a good ground plane and also to look amazing !
__________________
Al |
26th Aug 2016, 1:23 pm | #75 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: The deeper mysteries of 'ground'?
Different brass alloys differ in conductivity, you can reckon on 1/4 - 1/3 of the conductivity of copper, i.e. if you make the brass cross-section four times the equivalent copper conductor then you will be OK. Bronzes can be much lower, down to 1/10 of electrical copper.
Any connection made with an M3 screw is not going to be carrying very much current on the global scale of things. Once you are passing hundreds of amps through busbars of some feet in length, the difference starts to become very significant and brass b ecomes impractical. Depending on the frequencies of interest, the inductance of your interconnections might be much more significant than the ohmic resistance. |
31st Aug 2016, 3:16 pm | #76 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
Thank you, Lucien.
I remain unclear about whether to return the negative rail of the high voltage power supply I will construct should be returned to mains safety earth. It emerged from closer inspection of my high-grade lab power supply (previous post) that the -ve rail of that supply was indeed returned to safety earth. However, I have seen designs for high voltage power supplies where only the case of the isolating transformer and say, the metal case of the RFI suppression unit are returned to mains safety earth. My high -voltage power supply will be used for things like pulse experiments with a trigatron and so on. Only I will have access to it or operate it. It will contain several transformers whose laminations, at the very least, will be safety earthed , along with the case of the RFI suppressor . I note that it might be desirable to have a low impedance path for RF to earth, but in any case would like to distinguish what would be required from a regs standpoint of this were a commercial unit and what is desirable from an electronic viewpoint. Thank you
__________________
Al |
31st Aug 2016, 3:34 pm | #77 |
Guest
Posts: n/a
|
Re: The deeper mysteries of 'ground'?
It's always good to 'earth' the biggest bit of metal, at least you know where one bit is safe to touch. There is something to say about totally isolated equipment as if you only touch one bit it is OK, but with very high voltages even the smallest leakage current can lead to (at the very least) a surprise if not a dangerous shock. Some countries have the "isolated room" where no earth connexions are there, fine until you use an extension lead.
|
31st Aug 2016, 3:43 pm | #78 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: The deeper mysteries of 'ground'?
With HV power supplies you must consider what their applications will be that might require both ends floated, and to what voltage. If neither side is earthed, will they require equal insulation for symmetric operation? Or if you have a nominal 'high' end but your 'low' end is still floating, what would happen if the high end got earthed? Would the HV then appear across parts at the low end that would not withstand the voltage? Once you have established the operating envelope for everything that can be subjected to HV, you can identify whether it is necessary to force one end to earth, or near earth, to prevent damage.
Re your comment and previous about only yourself using it, making economies that significantly impact safety in what you design, simply because you don't anticipate anyone else using it, is suspect methodology. I cannot recall what economies or shortcuts I made in things 30 years ago, I am human, I forget, make mistakes, overlook things that were obvious at the time but not so at a later date. Self-imposed limitations on functionality might be reasonable but aim for commercial standards of safety. |
31st Aug 2016, 4:21 pm | #79 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: The deeper mysteries of 'ground'?
Quote:
Your more technical opening paragraph is very illuminating in terms of more rigorous and systemic thinking around circuit design. I know this is something you do for a living, whereas I do not. Hence it is a helpful conceptual tool, so many thanks for that, too.
__________________
Al |
|
7th Sep 2016, 3:00 pm | #80 | |
Hexode
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 407
|
Re: The deeper mysteries of 'ground'?
Quote:
|
|