You make valid points and it is the norm. Gold is used to prevent oxidation, but if its done poorly, the the base metal will emigrate through the gold. Yeah, I can see how a solder mask can alter the sound, it exist to prevent shortage and other contaminations on the tracks, so a pure protective layer.
Maybe a transparent solder mask is acceptable if I can locate one. And yes, there 101 things one can do to drive the price up. Snake oil effect at play ... he he
The more you talk about it, the more stupid the entire concept becomes. The existence of unicorns are more likely than electrons running away.
Thanks for the links, I will take a look at them. But I also think I found the PCB. The R03006 and R03010 is of the type I was looking for, so will read up on them. They have desirable properties. They do send out samples, so I can investigate the substrate at home.
RO3006? and RO3010? Laminates
I have done quite a lot of EMC work for some interesting projects (the sort you don't discuss) the expert used in these situation's as always been a proper rf engineer.
I have done many commercial projects, they just hired an 'EMC expert' who often had less knowledge of the problem than I did.
RF guys make good EMC experts but are expensive, 'EMC experts are cheaper but cost more.
I have done many commercial projects, they just hired an 'EMC expert' who often had less knowledge of the problem than I did.
RF guys make good EMC experts but are expensive, 'EMC experts are cheaper but cost more.
The small grain of truth could be that a point source is a lower impedance to air.
Look at the toroid on top of Van De Graff generators. (maintaining a charge)
Look at the reason why lightning rods work. (charge bleed off to prevent charge buildup)
A 90 degree corner is closer to a point source than 45's.
My $CDN 0.02
🙂
Solder mask will not alter the sound, at most it has a slight effect on the velocity factor of microstrip traces, this being very small and not critical at audio frequencies, or for most digital signals.
Again the Rogers is rather over the top and expensive for what you are going to design, to put it into perspective these esoteric high performance laminates barley account for 5% of ALL PCBs produced in the world.
Standard FR4 is much easier to handle and work with especially if you are etching at home.
If anyone has nay technical info as to the how and why of this solder resist myth I would like to see it (never had any explanation presented yet, but there is hope).
Again the Rogers is rather over the top and expensive for what you are going to design, to put it into perspective these esoteric high performance laminates barley account for 5% of ALL PCBs produced in the world.
Standard FR4 is much easier to handle and work with especially if you are etching at home.
If anyone has nay technical info as to the how and why of this solder resist myth I would like to see it (never had any explanation presented yet, but there is hope).
Last edited:
Different kettle of fish, and 90 degree corners DO NOT HAVE A SHARP POINT, they always have a radius. Electrons do not fall off at the corners of PCB traces.....Do not forget that electrons are only drifting in wires and traces at a very slow pace slower than 0.1mm/s when a signal is flowing (not talking about orbit velocity here) so are not even going fast enough to fly off the corner.
EMC/EMI and RF noise is a real thing yes, but not a problem at the level of design and development I am at, at the moment. Did I miss a point you are trying to forward ?
The post was a direct comment (and cynical responce) to DF96s post on EMC experts.
Why is EMC not a problem to the level you are working to? it does not care what level your are at or working to EMC is a problem for all designs, what many forget is EMC is a double sided coin EMI/signal integrity are the two sides...now rather than look at exotic materials you may be better considering these problems...mains hum is an EMC problem at the end of the day.
Why is EMC not a problem to the level you are working to? it does not care what level your are at or working to EMC is a problem for all designs, what many forget is EMC is a double sided coin EMI/signal integrity are the two sides...now rather than look at exotic materials you may be better considering these problems...mains hum is an EMC problem at the end of the day.
Allow me to clarify. I am in an early stage of development. I have speakers, pre & power amplifiers, D/A Converter among many things that I will develop and build. EMC, EMI and RF is something that I will address when time calls for me to focus on that. Em and RF shielding is something I take very seriously and will apply proper care for integration of such noise. Exactly what that mean or involve at the moment is something I've not looked at, but as you point out, is an important stage of the development.
Transformer noise, be that mains hum, EM stray fields or whatnot is also connected to how the parts are constructed and dampened. Not that I am an expert, far from it, but I know that aluminium and copper will do a good job at shielding this.
Anyone who has performed Millikan's oil drop experiment will have little doubt that electrons exist. Also anyone who has watched TV on a CRT set, or used a real oscilloscope.Oneminde said:
Of course, circuit theory and electromagnetism does not depend on the existence of electrons but the existence of electric charge. Even if electrons do not exist it would still be the case that smooth curves in PCB tracks are unnecessary below GHz frequencies. However, device physics does rely on electrons so you had better regard all components as 'black boxes' if you doubt the reality of electrons.
Audio PCBs rarely work at kV potentials. The 'grain of truth' is imaginary here. We really shouldn't try to legitimise nonsense by gracing it with postulated mechanisms.DUG said:
Just placed a sample order with Rogers. RO4*** is not PTFE, but a more cost effective high grade substrate while RO3*** is PTFE.
I can now see the picture in post one.
The blue stuff is nothing special, it is peelable resist.
Peelable Solder Mask
Also now I have seen the boards I would worry about their claim on the gold thickness, it would cause solder joint problems if it was as thick as claimed.
The blue stuff is nothing special, it is peelable resist.
Peelable Solder Mask
Also now I have seen the boards I would worry about their claim on the gold thickness, it would cause solder joint problems if it was as thick as claimed.
Last edited:
In regards to your post, I found this:
RECOMMENDED PCB PLATING FOR DOME APPLICATIONS
(Proper plating is essential to prevent deterioration of the circuit pad over time)
NICKEL
Electro-Deposited Nickel, Bright Hard
<1,000,000 cycle life: 0.00005” – 0.0002” thick (50-200 micro inches or 1.25 – 5.0 microns)
>1,000,000 cycles life: 0.0002” – 0.0005” thick (200-500 micro inches or 5.0 – 12.5 microns)
GOLD
Electro-Deposited Gold, Hard (Ref – MIL-G-45204C, Type II, Grade C)
<1,000,000 cycles life: 0.000015” – 0.00003” thick (15 – 30 micro inches or 0.38 – 0.76 microns)
>1,000,000 cycles life: 0.00003” – 0.00005” thick (30 – 50 micro inches or 0.76 – 1.27 microns)
PCB Plating for tactile metal domes
and now considering:
"Leaf metal varies from 1/8000 mm to 1/10,000 mm in thickness." 0.1–0.125 µm
"Pure gold that is pounded into sheets applied to other surfaces by hand. Usually about 3 micro-inches thick." 0.08 µm
Meaning, 100 µm (0.1 mm) = about 100 gold leafs which is a redicilous amount. One need less than 1 µm to have a protective layer anyway and why not create an inverted solder mask while plating ? Meaning that the area you want to solder to is not covered in gold, only Tin - on a pre tinned board before gold. Inverted solder mask to cover sections that will be soldered is an interesting idea - at least experimental idea. The adhesiveness of the solder (PB free) should not cause any problems, how can it if the larger portion of the solder is adhered to Tin and just a small portion (overlapping) to the gold.
Last edited:
polyflon microwave
POLYGUIDE
Dielectric Constant (Cu Clad) 2.320 +/-.005 10GHz
Dissipation Factor 0.0005 10GHz
You can get from ebay sometimes
regards
james
POLYGUIDE
Dielectric Constant (Cu Clad) 2.320 +/-.005 10GHz
Dissipation Factor 0.0005 10GHz
You can get from ebay sometimes
regards
james
That is hard gold, you cannot solder reliably on hard gold. ENIG is used on boards that have to be soldered. Hard gold plating is used for connectors and on PCB switch contacts, it is wrong for soldering a board it leads to brittle joints if you can get the solder to wet.
Electroless nickel immersion gold - Wikipedia, the free encyclopedia
As you are new to this I would start of with basic PCB material especially for your first off, that may not work properly.
Then when you have learned more and have some experience then look at more esoteric PCBs (though sound wise there will be no gains).
Again that material is for microwave designs, not the best choice, basic good quality FR4 will give you the best results.
Last edited:
I am grateful for the advices. That µm list were a mere comparison to Luxman's claim about 100 µm. There is as you mention hard gold and soft gold. PCB traces, as I am learning, should have the soft gold plating which is also known as electrolytic gold aka Flash gold.
- Flash Gold: 5 –15 µinches (0.127 - 0.381 µm) = Solderable
- ASTM-B-488, Type III, Code A (Soft Gold)
- Electrolytic Gold: Electrolytic gold offers high electrical conductivity, solder-ability, weld-ability, infrared-reflectivity, and excellent corrosion resistance (as well as sealing the nickel surface to eliminate passivation). Electrolytic gold is primarily used in the electrical industry for RF connectors and printed circuits, and in the electronics industry for transistor integrated circuits.
This should take care of the gold issue. When it comes to the substrate issue, that is now more or less solved and I will write a new post on that. FR4 will work during a development stage.
This tread has now been solved.
I started this tread with a question if anyone knew what type of PCB Luxman and Accuphase use, more specifically Luxman. Not only did I get help in that, but this tread also threw me into debates about the shape of the tracks on the PCB, solder mask vs no solder mask, gold plating vs no gold plating, and EMI/EMF as well as RF shielding. This is ofc important and interesting topics, but it did not have anything to do with my initial question. Learning more than what one asked for is often a good thing, like here.
Lets get back on top. PCB substrate. After several days of research and indeed, very few clues towards the type Luxman is using (at least on some of their products), via a tip on Rogers Corporation's pcb's and range of substrates, I stumbled upon two types that is interesting. One being a PTFE (RO3000 series) type which Luxman claim their PCB is made of and another which is Glass-reinforced hydrocarbon and ceramic dielectric (RO4000 series). The thing is that the RO3000 which is a PTFE (Teflon) is not a semi transparent substrate as the RO4000 is, and PTFE is a non transparent plastic, same as Nylon is which is a similar material, same goes for Delrin.
RO4000® SERIES HIGH FREQUENCY CIRCUIT MATERIALS DESCRIPTION:
Woven Glass Reinforced, Ceramic Filled Thermoset
Rogers RO4000® high frequency circuit materials are glass-reinforced hydrocarbon/ceramic laminates-not PTFE. They combine high frequency performance comparable to woven glass PTFE substrates with the ease-and hence low cost-of fabrication associated with epoxy/glass laminates. The RO4000 advanced circuit material is a woven glass reinforced, ceramic filled thermoset laminate with a very high glass transition temperature (Tg >280°C). Unlike PTFE based microwave materials, no special through-hole treatments or handling procedures are required. RO4003C was designed with a dielectric constant of 3.38. The RO4350B has a dielectric constant of 3.48 and is UL94V-0 rated.
RO4450B™ and RO4450F prepregs were developed to enable users to build high frequency multilayers and are based upon the RO4000 series core materials. RO4450B and RO4450F prepregs are compatible in multilayer constructions with either RO4003C™, or RO4350B™ laminates. RO4000 series high frequency prepregs are compatible with the majority of standard FR4 processing practices. Prepreg technical information is provided in the RO4400 data sheet and fabrication guidelines below.
RO4500™and RO4730™ LoPro™ antenna grade laminates extend the RO4000 product series into large volume antenna applications. The ceramic filled, glass reinforced, hydrocarbon based material provides the controlled dielectric constant, low loss performance and excellent passive intermodulation response required for mobile infrastructure microstrip antenna applications. Contact us at 480-961-1382 for more information and sampling of these new products.
LoPro™ Reverse-Treated Copper Foil Option is now available for RO4000® series laminates! This special interface technology works with RO4000 materials for improved insertion loss and outstanding passive intermodulation characteristics.
Product Features:
- An alternative to PTFE
-Excellent high frequency performance due to tight dielectric tolerance and low loss
-Stable electrical properties versus frequency
-Low thermal coefficient of dielectric constant
-Low Z-axis expansion
-Low in-plane expansion coefficient
-Excellent dimensional stability
-Volume manufacturing process
-Easy, low cost fabrication
-RoHS compliant, environmentally friendly
They also mention that the fabrication is similar to FR4 which mean that is a more economical solution than PTFE. Also, by looking at the pictures, we can see that the Luxman and Rogers PCB look very similar and after all, that was my target - the visual appearance of the PCB. Upgraded properties compared to FR4 is a bonus.
http://integratedtest.com/products/pcb-fabrication/laminates/Rogers4003-4350.pdf
http://www.rogerscorp.com/acm/producttypes/9/RO4000-Laminates.aspx
So there we go. I solved it with a bit of help. So thank you diyAudio for the support.
Rogers confirmed my two samples for the RO4000 & RO3000 series which will allow me to check both types. I will post results for this.
An externally hosted image should be here but it was not working when we last tested it.
I started this tread with a question if anyone knew what type of PCB Luxman and Accuphase use, more specifically Luxman. Not only did I get help in that, but this tread also threw me into debates about the shape of the tracks on the PCB, solder mask vs no solder mask, gold plating vs no gold plating, and EMI/EMF as well as RF shielding. This is ofc important and interesting topics, but it did not have anything to do with my initial question. Learning more than what one asked for is often a good thing, like here.
Lets get back on top. PCB substrate. After several days of research and indeed, very few clues towards the type Luxman is using (at least on some of their products), via a tip on Rogers Corporation's pcb's and range of substrates, I stumbled upon two types that is interesting. One being a PTFE (RO3000 series) type which Luxman claim their PCB is made of and another which is Glass-reinforced hydrocarbon and ceramic dielectric (RO4000 series). The thing is that the RO3000 which is a PTFE (Teflon) is not a semi transparent substrate as the RO4000 is, and PTFE is a non transparent plastic, same as Nylon is which is a similar material, same goes for Delrin.
RO4000® SERIES HIGH FREQUENCY CIRCUIT MATERIALS DESCRIPTION:
Woven Glass Reinforced, Ceramic Filled Thermoset
Rogers RO4000® high frequency circuit materials are glass-reinforced hydrocarbon/ceramic laminates-not PTFE. They combine high frequency performance comparable to woven glass PTFE substrates with the ease-and hence low cost-of fabrication associated with epoxy/glass laminates. The RO4000 advanced circuit material is a woven glass reinforced, ceramic filled thermoset laminate with a very high glass transition temperature (Tg >280°C). Unlike PTFE based microwave materials, no special through-hole treatments or handling procedures are required. RO4003C was designed with a dielectric constant of 3.38. The RO4350B has a dielectric constant of 3.48 and is UL94V-0 rated.
RO4450B™ and RO4450F prepregs were developed to enable users to build high frequency multilayers and are based upon the RO4000 series core materials. RO4450B and RO4450F prepregs are compatible in multilayer constructions with either RO4003C™, or RO4350B™ laminates. RO4000 series high frequency prepregs are compatible with the majority of standard FR4 processing practices. Prepreg technical information is provided in the RO4400 data sheet and fabrication guidelines below.
RO4500™and RO4730™ LoPro™ antenna grade laminates extend the RO4000 product series into large volume antenna applications. The ceramic filled, glass reinforced, hydrocarbon based material provides the controlled dielectric constant, low loss performance and excellent passive intermodulation response required for mobile infrastructure microstrip antenna applications. Contact us at 480-961-1382 for more information and sampling of these new products.
LoPro™ Reverse-Treated Copper Foil Option is now available for RO4000® series laminates! This special interface technology works with RO4000 materials for improved insertion loss and outstanding passive intermodulation characteristics.
Product Features:
- An alternative to PTFE
-Excellent high frequency performance due to tight dielectric tolerance and low loss
-Stable electrical properties versus frequency
-Low thermal coefficient of dielectric constant
-Low Z-axis expansion
-Low in-plane expansion coefficient
-Excellent dimensional stability
-Volume manufacturing process
-Easy, low cost fabrication
-RoHS compliant, environmentally friendly
They also mention that the fabrication is similar to FR4 which mean that is a more economical solution than PTFE. Also, by looking at the pictures, we can see that the Luxman and Rogers PCB look very similar and after all, that was my target - the visual appearance of the PCB. Upgraded properties compared to FR4 is a bonus.
http://integratedtest.com/products/pcb-fabrication/laminates/Rogers4003-4350.pdf
http://www.rogerscorp.com/acm/producttypes/9/RO4000-Laminates.aspx
So there we go. I solved it with a bit of help. So thank you diyAudio for the support.
Rogers confirmed my two samples for the RO4000 & RO3000 series which will allow me to check both types. I will post results for this.
Last edited:
- Status
- Not open for further replies.