Frank was asking about H field emissions - these would be right down the bottom of the list of worries for a PC.
And hitswire was talking about the switchmode supply in a PC being a source of those emissions.
se
I think you will find that in the typical home invironment it is as i said. E field dominate at >100Khz-1Mhzand above, while H fields are generally much below that range.
An issue for audiophiles -- the prolific SMPS can produce both field coupled and direct coupled noise. EMI/RFI government specs from equipment is typically directed to aircraft and emergancy, police et al frequencies and are tested to comply by agencies such as USA's UL. However, those tests and freqs are far above the audio range and thus audiophiles need a lot better performance than UL qualified equipment provides. And, they dont cover the direct-coupled issues into the ac power lines. Thx-RNM
Last edited:
Very much so, home and woik.
Finishing up a bathroom reno, so my clock stuff is on hold. Doing the radiant heat next.
Exactly. Moreso PC's though, Apple really paid close attention to EM radiation.
Actually, in the HF range, it's E/M, propagating waves at 377 ohms that is a biggie. House wiring and line cords are long enough to couple to the air to launch the waves.
Ground loops will also help M fields launch.
E fields are generally the easiest to stop radiating. M fields are the hardest.
jn
What was funny: Apple industrial designers had a strong influence on the JBL Soundsticks three-piece system, which featured transparent housings for everything and initially only a USB input. The Micronas USB DAC was promised to be low emissions according to Apple. Well, it wasn't, except in a standby mode, and a good deal of time had to be spent in the EMC lab to render the product compliant and still allow everyone to stare at the circuit boards. There was a fight over the ferrite doughnut in the input cable, much wringing of hands and wailing from ID over its inclusion.
The product was quite successful. Some mistakenly imagined that it was USB-powered since around that time a patent of mine issued for managing such (6178514). Another engineer supposed that the four drivers per satellite were driven in some sort of complex phase relationship (and warned us about Bose patents, which he was never able to produce
). But no, it used an a.c. adapter, and the drivers were acoustically in parallel while alternating in magnetic polarity to partially cancel adjacent monitor distortions.At one point, since people were frustrated that they couldn't use analog sources, Plunkett changed the design to an analog-input version and added touch controls for volume to a redesigned satellite base.
Nelco 4013-SI and Panasonic Megtron 6 are pretty much at the top these days. Rogers popularity for very high performance applications has declined, particularly with the introduction of the newer Megtron 6. Both Nelco 4013 and Megtron 6 are robust to the higher reflow temperatures of ROHS 6/6 lead-free processes, while most Rogers is marginal in that regard. Rogers is playing catch-up now. Nelco 4013 and Panasonic Megtron 6 are both less expensive than Rogers. All specs that I am aware of for Megtron 6 are the equal of Rogers to 10 GHz. Most systems in which 28 Gb/s signals are routed on copper traces utulize Megtron 6 nowadays.
Cheers,
Bob
Good quality FR4 would do the job, PC boards aint up in the high GHz region so you dont realy need exotic materials and probably wont get much benefit out of them. What you need to make a PC motherboard design as quiet as possible is a good layer stack up, and good layout and routing, especialy the SMPS. Also decoupling properly using power integrity simulation software helps. The main power plane layers should be closely coupled, and here is one area where a better dialetric would pay of, to maximise he planar capacitance as this provides the instantaneous switching supplies, long before any decoupling capacitors.Standard PC motherboards are layed out with the minimum number of layers to cut costs, quite often 6-8, for signal integrity and noise suppression they should have at least 12. But the best layout technique is HDL, plenty of info in this if you realy want to create the quietist design:
The HDI Handbook
A lot of consumer products (apple included) use this for of layout with the smallest possible packages, so you get the smallest loops, in built shielding with HDI and the right number of layers for signal integrity.
There is no free meal with PCB design, using more exotic materials adds other problems to your design criteria.
The HDI Handbook
A lot of consumer products (apple included) use this for of layout with the smallest possible packages, so you get the smallest loops, in built shielding with HDI and the right number of layers for signal integrity.
There is no free meal with PCB design, using more exotic materials adds other problems to your design criteria.
Slightly OT:
If any of you guys plan to come to Denver this weekend for the Rocky Mountain Audio Fest, be sure to stop by our little audio diy booth in the lobby (behind the registration counter).
We'll have a shared booth from Elektor USA, AudioXpress and Linear Audio.
Bob Cordell will be there showing his VinylTrak phono/preamp.
Be sure to stop for a light chat or serious diyaudio business
We'll be there the whole show.
jan didden
If any of you guys plan to come to Denver this weekend for the Rocky Mountain Audio Fest, be sure to stop by our little audio diy booth in the lobby (behind the registration counter).
We'll have a shared booth from Elektor USA, AudioXpress and Linear Audio.
Bob Cordell will be there showing his VinylTrak phono/preamp.
Be sure to stop for a light chat or serious diyaudio business
We'll be there the whole show.
jan didden
Slightly OT:
If any of you guys plan to come to Denver this weekend for the Rocky Mountain Audio Fest, be sure to stop by our little audio diy booth in the lobby (behind the registration counter).
We'll have a shared booth from Elektor USA, AudioXpress and Linear Audio.
Bob Cordell will be there showing his VinylTrak phono/preamp.
Be sure to stop for a light chat or serious diyaudio business
We'll be there the whole show.
jan didden
I'm really looking forward to seeing Jan and all the other folks who are fortunate enough to make it to this great audio show. My friend Peter Smith will also be there.
For those of you interested in the Linear Integrated Systems line of JFETs, I believe Paul Norton will also be at RMAF.
Cheers,
Bob
I would like to follow up on board material. Almost 40 years ago, Tektronix found a problem with FR board material. They had trouble making resistive attenuators because the board material had a significant amount of dielectric absorption. They called this problem, 'circuit hook'.
At Vendetta Research, about 25 years ago, we started with FR-4, but then we went to a 'better' material, (certainly more expensive, with better specs) and we got a subjective improvement. Then when we made the CTC Blowtorch boards, we went to Teflon. Each small board then cost us about $50ea in lots of 100, to make, but I am not sorry that we did so.
Today, FR-4 is OK for most applications, but I still hope for better. Is Teflon worth it? I am not sure, but it sure didn't hurt us in achieving our goal.
I now have the same questions about Polystyrene and Teflon. Is one really better that the other for audio applications? I really don't know for sure.
Most mid-fi seems to use something cheaper and smellier than FR-4. I would not use it for hi end, myself.
At Vendetta Research, about 25 years ago, we started with FR-4, but then we went to a 'better' material, (certainly more expensive, with better specs) and we got a subjective improvement. Then when we made the CTC Blowtorch boards, we went to Teflon. Each small board then cost us about $50ea in lots of 100, to make, but I am not sorry that we did so.
Today, FR-4 is OK for most applications, but I still hope for better. Is Teflon worth it? I am not sure, but it sure didn't hurt us in achieving our goal.
I now have the same questions about Polystyrene and Teflon. Is one really better that the other for audio applications? I really don't know for sure.
Most mid-fi seems to use something cheaper and smellier than FR-4. I would not use it for hi end, myself.
- Status
- Not open for further replies.