PCB substrate question (Teflon Luxman & Accuphase oriented)

[Oneminde]

[SOLVED] PCB substrate question (Teflon Luxman & Accuphase oriented)

********** This tread is now solved************
See post #36 - http://www.diyaudio.com/forums/cons...-luxman-accuphase-oriented-4.html#post3980922

Hello fellow diyAudio members. This is my first post here.

My question is oriented around the pcb substrate Accuphase and Luxman is using. Both state on different places, that they use the PTFE type which is off-white type. I stumbled upon 3 articles about Luxman and I am a bit stuck in understanding everything involved, so maybe someone here can guide me.

Luxman

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

Here is what the articles write:

"My favorite item in the overall design of the M800A is the use of specially fabricated, gently curved signal paths. All primary signals in the amplifier’s circuitries traverse on these specially printed routes with curved turns that resemble miniature race tracks. The PCB on which these beautiful signal paths are pressed is of specially prepared peel-coat type that is peeled off once the traces are printed. This PCB is then further coated with a 100µm-thick gold-plating to preserve the low-impedance and frequency response characteristics of the signal path against corrosion. (Note: “µm”, micrometer. 1,000 µm=1mm.)"

and

"Printed Circuit Boards normally have a green coat (dye) that can be printed upon, so the manufacturer knows where to place the copper traces and where to insert the resistors & capacitors. Luxman's designers don't like the sound of the green dye, in their extensive research they found that it smears the sound, so instead they print onto a plastic peel coat and remove it once the traces and mounting holes are in place. Then the copper traces are gold plated to reduce resistance across the circuit, as well as eliminate copper's corrosion over time.."

My two questions are as follow:

1. If I order a PTFE substrate PCB, will then get this off-white colour ? for it seams as there are types that have a brownish surface and I do not want that.

2. What is the purpose of the peel-coat. I am not talking about the photo etching resist and since Accuphase and Luxman is not using a solder mask after the gold plating. So I am kinda lost here. Any help ?

 

[marce]

They are sprouting a load of BS...
Go for FR4 standard PCB, 45 degree traces and use a solder resist (that is what the green stuff is) you will not notice or measure any difference in your signals.
The gold wont have any affect other than stopping tarnishing (not a problem anyway) and has a lower conductance than copper so will not reduce resistance.
Don't know what curved tracks are going to achieve maybe stops the electrons falling off at the corners....
Plenty of real information out there regarding PCBs, without resorting to esoteric audiophile nonsense.
Cant see the pictures by the way, a better link and one to the text would be nice....

 

[Neutrality]

I agree.

 

[Oneminde]

They are sprouting a load of BS...
Go for FR4 standard PCB, 45 degree traces and use a solder resist (that is what the green stuff is) you will not notice or measure any difference in your signals.
The gold wont have any affect other than stopping tarnishing (not a problem anyway) and has a lower conductance than copper so will not reduce resistance.
Don't know what curved tracks are going to achieve maybe stops the electrons falling off at the corners....
Plenty of real information out there regarding PCBs, without resorting to esoteric audiophile nonsense.
Cant see the pictures by the way, a better link and one to the text would be nice....

Here are the two articles.

1. On a Higher Note :: M-900u
2. A Luxman heritage illustration and a review of the $16,000 M-800A 80th Anniversary stereo power amplifier Review - Equipment Reviews - Dagogo

There are two reason why. The look of off-white PCB - should be PTFE according to my knowledge. And gold plated, and I will not need a solder mask.

I am thankful for the reply, but it did not really answer my question on Teflon or the purpose for the peel-coat, so still equally lost.

I get that there exist audiopholery and while some of that makes sense, not everything does, so I ask and I try to learn. What is obvious for someone skilled in the art of electronics may not be as obvious for someone ells, and there is the matter of taste and what looks good etc.
There are people that are happy with the project existing inside a shoebox, I am not that person.

 

[SY]

If your esthetic sense demands a white circuit board, I would bet that this is available with materials other than PTFE. Besides needless expense for audio applications, PTFE boards have markedly poorer mechanical properties and can be quite unreliable compared with more conventional epoxy resins.

 

[DF96]

High-end audio uses high-mobility electrons which travel faster than the ordinary electrons used by lesser equipment. This means that they need a larger bend radius; otherwise they tend to accumulate at the corners of the PCB traces and create smearing of the stereo image. This problem can to some extent be reduced by ensuring that the tracks for the two stereo channels are identical - a technique used for years in high-speed digital circuits but only recently applied to audio.

These techniques (along with gold-plating) produce beautiful circuits and, as everyone knows, beautiful circuits create a beautiful music experience. In order to get the best advantage from this it is of course important that your audio sources and ancillary equipment are of the highest calibre.

 

[kevinkr]

Teflon pc boards are more flexible than FR4 which means some care in handling is required. It has a lower and more stable dielectric constant which is relevant for microwave applications, but is otherwise probably more trouble than it is worth. Leakage is lower which is relevant in some very high precision analog measurement hardware. (semiconductor ATE and similar)

Again the lack of sharp angles is a technique useful in microwave and high voltage circuitry.

None of this hurts, but whether or not it is of any particular benefit is another question.

Speculating, but I assume the peel coat protects the board from damage during shipping and assembly. Possibly it could also be used during drill, and replaced subsequently. Presumably during assembly it performs the same function as the solder mask.
What are you building that requires this level of care?

 

[marce]

Plenty of esoteric materials here:
Advanced Circuit Materials - Rogers Corp

How often are they used in real life, very rarely and only on the very top 5% of designs...
Cant answer the Luxman quotes further because they are meaningless and this:

Printed Circuit Boards normally have a green coat (dye) that can be printed upon, so the manufacturer knows where to place the copper traces and where to insert the resistors & capacitors. Luxman's designers don't like the sound of the green dye, in their extensive research they found that it smears the sound, so instead they print onto a plastic peel coat and remove it once the traces and mounting holes are in place. Then the copper traces are gold plated to reduce resistance across the circuit, as well as eliminate copper's corrosion over time..

is written by someone who has no idea about PCBs.
Interestingly I was discussing sharp corners the other day at work (90 degree bends don't even give a really sharp corner, 45s are not sharp at all)as they have no real affect on digital up to 2GHz....

As to FR4 there are many types the generic name covers a whole host of temperature ranges and weave patterns, from basic cheep stuff to more exotic FR4 for very high speed designs etc.
The only time I don't use solder mask on boards is when they are for the European space agency or similar, in space they don't like solder mask.....

 

[Oneminde]

Teflon pc boards are more flexible than FR4 which means some care in handling is required. It has a lower and more stable dielectric constant which is relevant for microwave applications, but is otherwise probably more trouble than it is worth. Leakage is lower which is relevant in some very high precision analog measurement hardware. (semiconductor ATE and similar)

Again the lack of sharp angles is a technique useful in microwave and high voltage circuitry.

None of this hurts, but whether or not it is of any particular benefit is another question.

Speculating, but I assume the peel coat protects the board from damage during shipping and assembly. Possibly it could also be used during drill, and replaced subsequently. Presumably during assembly it performs the same function as the solder mask.
What are you building that requires this level of care?

To everyone, thank you for the reply's.

******
@ kevinkr

Your explanation for the peel coat makes sense since that is what I consider it to be used for as well, protection during storage and transport. I guess that is important for equipment that cost several thousands of $. Not that the value is connected to parts cost, something which buggers me quite allot.

Highly rated audiophile equipment is over rated in terms of value and quite often the price level is just stupid. Lets take an example. There exist a vinyl record player that use the same type of electro motor that can be found on the Mars Rover. This vinyl record player costs $125 000,- and that is just plain stupid. Then we have the low end, low cost components and electronics that barely do what they are suppose to do with bad to generic sound. Some people will be okay with that and do not really have any requirements besides the speakers should make noise. I exist in-between these two levels. I want high grade equipments, but I do not want to pay a small fortune - if that makes sense as I am sure it does for many diy people, why ells put in time and money it is required to produce equipment. Sure, for the fun of it works, but that can hardly be the only reason.

I am planing several parts in a Music & Home Cinema solution and I am inspired by Tidal Audio, B&W, Morten Design, Accuphase, Luxman, Marantz and some of these use among other Duelund Coherent Audio, Accuton and so forth. But the cost of equipment from these producers - using Accuton, Duelund, Nichicon cap's etc is just a big no. However, develop and construct my own line of pre-amps, power amps, speakers is something I can manage. So I want quality, high grade and God damned it, good looking stuff, on a budget.

Hope that answer the question ?

 

[Oneminde]

Plenty of esoteric materials here:
Advanced Circuit Materials - Rogers Corp

How often are they used in real life, very rarely and only on the very top 5% of designs...
Cant answer the Luxman quotes further because they are meaningless and this:

is written by someone who has no idea about PCBs.
Interestingly I was discussing sharp corners the other day at work (90 degree bends don't even give a really sharp corner, 45s are not sharp at all)as they have no real affect on digital up to 2GHz....

As to FR4 there are many types the generic name covers a whole host of temperature ranges and weave patterns, from basic cheep stuff to more exotic FR4 for very high speed designs etc.
The only time I don't use solder mask on boards is when they are for the European space agency or similar, in space they don't like solder mask.....

The reduction of resistance is BC, that is a fixed material value, the gold prevents Cu oxidation and that bears merit and the only thing of interest besides that it looks nice - I am a sucker for design and how things look yes, but that does not mean I am stupid.

Now see, the part about different grades of FR4 is important and since you know what you are talking about. What type / grade should I focus on if I want the same look as the Luxman PCB ?. I guess epoxy fibreglass FR4 is fine if the substrate is white - maybe a pure design / look aspect if the substrate does very little in terms of colouring the function.

In regards to sharp and non sharp corners, 45-degree is the standard besides if there is no room to do that. Many use 90-degree and that works fine as well, but soft corners are more gentle on the signal. Its not like I think that a 90-degree corner will throw electrons out from the path 🙂

 

[irribeo]

I agree, so are the pins and connectors curved too? What about all those holes in that Luxman pcb, curved as well? I wouldn't consider Luxman if not every part used was internally and externally curved too.

 

[Oneminde]

I agree, so are the pins and connectors curved too? What about all those holes in that Luxman pcb, curved as well? I wouldn't consider Luxman if not every part used was internally and externally curved too.

Pun intended yes ? 🙂

 

[mondogenerator]

I'm sure you can get white soldermasked FR boards made....with up to 4oz (100um?) copper. Tinned is fine enough.

 

[Oneminde]

I'm sure you can get white soldermasked FR boards made....with up to 4oz (100um?) copper. Tinned is fine enough.

Yes, white solder mask is available as well as electroplating Cu onto it. I am not planing to have solder mask on it, only Au on the Cu tracks and painted (silk screen) the symbols and letters and leave it that way. Less is more. 🙂

 

[marce]

Use a solder mask, apart from going into space there is no reason not to and many reasons to. If you don't want to use a solder mask (!) you can always have the whole design finished with ENIG, cheaper than a hard gold coating. But as most boards in the world are standard solder mask finish I would go with that.
Even though I mentioned 90 deg corners I do not use them I use 45 deg corners and on some rf boards I use curved traces for the rf feeds.
A high Tg FR4 would be a good start, 170deg. (Tg is glass transition temperature).

Some links on 90 deg corners...
The technical part:

http://fruttenboel.verhoeven272.nl/pads/data/mitering.pdf

Who's Afraid of the Big Bad Bend?

http://emcmadesimple.com/pdfs/publications/corners-USA.pdf



The fun part:

OF FLYING ELECTRONS

Article from above, look for routing:

EMC Design for High Speed PCB's

 

[DF96]

Its not like I think that a 90-degree corner will throw electrons out from the path

Of course you don't think that.

Many use 90-degree and that works fine as well, but soft corners are more gentle on the signal.

Or maybe you do?

Even among those of us who know that nonsense is nonsense there can sometimes be a residual feeling that maybe there is a grain of truth in it. We then need to remember our scientific education and firmly remind ourselves that nonsense truly is nonsense however popular it is. Science tells us that the shape of PCB track corners may matter electrically from a few GHz upward. Below that it is purely cosmetic, or possibly a matter of mechanical issues during fabrication.

 

[marce]

DF96 have a read at the of flying electrons link, its from some discussions around 96 prompted by the link below it and specifically this statement (from an EMC expert 🙁)

Routing

Orthogonal trace corners should be avoided. The debate rages, but as frequencies and edge rates continue to rise, ninety degree corners introduce excess capacitance and cause a small change in characteristic impedance. This becomes disastrous at high frequencies (e.g. 100 MHz) when electrons virtually fly off the sharp corners of the bend. Forty-five degree turns, with a minimum segment length of twice the trace width, are better. Arced corners, with the inside radius of at least the trace width, are by far the best approach for high-speed signals.

Avoiding 90 deg corners is down to the etching process, in't old days it was not as well controlled as it is today and acid traps were a big concern (the board was hung in a bath of etchant), these days due to deigns sizes better techniques exist such as spray etching. Also you can get more traces into a given area using 45 deg corners.

 

[marce]

Yes, white solder mask is available as well as electroplating Cu onto it. I am not planing to have solder mask on it, only Au on the Cu tracks and painted (silk screen) the symbols and letters and leave it that way. Less is more. 🙂

Less is not more, solder mask is best, it also helps in soldering and avoiding shorts. Gold is never directly plated onto copper, there is always a nickel coating and sometimes a nickel and a palladium layer...the other problem with gold is embrittlement of the solder joint and contamination of your solder, depending on the soldering method used (wave, selective, hand reflow etc).
The belief that solder mask alters the sound is truly delusional and ranks up there as the most stupid AP belief.
Skin effect also does not come into play at audio frequencies so the gold serves no other purpose than putting up the price.

 

[DF96]

Many years ago when I worked with computers there appeared a new breed of IT person called a 'network manager'. Talking to these people I discovered that many of them seemed to know less about networks than I did (e.g. had no idea of Ethernet packet format, would not recognise a routing algorithm if it landed on their foot) yet they were very highly paid as they had learnt a few network management commands. They were much in demand. They were very good at setting up fast but unreliable LANs.

I guess the modern emphasis on at least paying lipservice to EMC means that 'EMC experts' are now highly sought-after. Inevitably this will attract people who don't actually understand EM but have learnt a few words and a few rules-of-thumb. I assume this explains how an 'EMC expert' can talk of electrons "virtually flying off the corners" at a mere 100MHz. It may also explain why so many modern items have extremely poor EMC performance.

 

[Oneminde]

Of course you don't think that.

Or maybe you do?

Even among those of us who know that nonsense is nonsense there can sometimes be a residual feeling that maybe there is a grain of truth in it. We then need to remember our scientific education and firmly remind ourselves that nonsense truly is nonsense however popular it is. Science tells us that the shape of PCB track corners may matter electrically from a few GHz upward. Below that it is purely cosmetic, or possibly a matter of mechanical issues during fabrication.

Yeah, contradicting myself there. The existence of electrons as a physical particle is more theoretical than physical and is a hypothetical conceptualisation of how the world works. Traces of a behaviour is not equal to proof. CERN capture traces of particles, all they see is that something were there and can then calculate different theoretical maths about how they think the particle is. But no real observations can be done .. yet.

Some people are extremely dogmatic about the existences of particles as there are dogmatists about the need for a deity for the existence of the universe. What we do know is that a force is moving in close proximity of a conduit. Be that particle force or aether force, non the less, we can manipulate nature to do things fore us and in most cases it works. So I guess I am an electrical🙂 agnostic as well