If you want to get a grip of the whole aspect of PCB design, fabrication and assembly then look at the IPC, they produce guidelines, specifications and standards for all aspects. If you want to choose a base laminate then there is a host of documentation covering ALL the laminate types from the organic based to the inorganic, there are many different weave patters for glass reinforced laminates, covering basic design to very high speed, where the impedance discontinuities over the weave are minimised (1067 is a good example).
This is a list of the basic IPC specifications, for all commercial designers you should be specifying these when getting boards manufactured... The 40XX series covers the wide range of board materials, weave patterns etc. All have their uses, for audio there is absolutely no reason to not use one of the many types of FR4...
https://portal.ipc.org/Purchase/ProductDetail.aspx?Product_code=4499ABA8-7C32-DD11-A3F8-001422202D38
As to bad PCB designers as I stated yesterday there are many who think PCB design can be done by anyone and there are those who think doing a basic course in a PCB design package qualifies them... Not so, if you want to have an idea you are picking a designer who really is a PCB designer look for someone who is CID or preferably CID+, I would like to say all PCB designers have the same dedication to the craft, but they don't and that and the general consensus that its a skilless vocation does not help.
List of certified designers....
https://portal.ipc.org/Purchase/ProductDetail.aspx?Product_code=4499ABA8-7C32-DD11-A3F8-001422202D38
This is a list of the basic IPC specifications, for all commercial designers you should be specifying these when getting boards manufactured... The 40XX series covers the wide range of board materials, weave patterns etc. All have their uses, for audio there is absolutely no reason to not use one of the many types of FR4...
https://portal.ipc.org/Purchase/ProductDetail.aspx?Product_code=4499ABA8-7C32-DD11-A3F8-001422202D38
As to bad PCB designers as I stated yesterday there are many who think PCB design can be done by anyone and there are those who think doing a basic course in a PCB design package qualifies them... Not so, if you want to have an idea you are picking a designer who really is a PCB designer look for someone who is CID or preferably CID+, I would like to say all PCB designers have the same dedication to the craft, but they don't and that and the general consensus that its a skilless vocation does not help.
List of certified designers....
https://portal.ipc.org/Purchase/ProductDetail.aspx?Product_code=4499ABA8-7C32-DD11-A3F8-001422202D38
I do have to disagree regarding the creativity, the best PCB designers are creative. They produce the neatest tightest and best designs both in appearance and functionality... We had an example posted on here a few weeks ago.
Here is a basic PCB manufacturing specification that illustrates the IPC specifications your boards will be manufactured too (some text would need changing as its my generic spec and gets altered on the fly for different organisations) what it does is show there is a greater depth to PCBs and specifying the bare boards than many seem aware of....
We had excellent results with FR4 in our 10 GBit/s fiber optic transceivers, and
you see the slightest errors directly in the eye diagram there. The MSA standard
calls for direct board edge connectors, already that would have been impossible
with teflon because of its inferior mechanical properties.
And we had Kapton flex connections to the laser and pin diode assemblies,
now that was a major problem with the transitions and the, yes, flexibility.
We spent a lot of time with Agilents Advanced Design System and Ansoft
until we had a repeatable solution.
Building GHz stuff on FR4 is actually quite easy if you do not need resonant structures.
Of course, you cannot fabricate that in a Chinese backyard company. But
laminate producers like Isola are capable of specifying their stuff and deliver
to the specs.
And remember, a GHz is a million KHz, and if an AC problem can be handled up there,
it is a million times easier at nervous DC. So you can be pretty sure that the necessity
of Teflon boards at AF is the invention of a high end salesdroid.
Gerhard
Last edited:
A good point, the more exotic laminates are not only more fragile, but it is also harder to manufacture the finished board so you pay a lot of money for NO added benefit....
When you are talking about wavelengths in the low single digit mm, its no wonder connector geometry and mating surface gaps etc are important. Pico Henries is a lot of L at 10 GHz.
Exactly, and as long as you have control of the laminate to get repeatability.
That can be tricky with Chinese CEMs.
Definitely!!
Actually, there's disadvantages to materials like PTFE for audio applications, but if you're selling fashionable voodoo, that's what the dupes want.
You can tell me about electrical effects of screws loosening, traces de-laminating, boards flexing with vibration.
to be fair, glue it properly to a 1cm thick aluminium ground plane then bolt it into a machined from solid box and it'll survive the 11G of launch, but you make several to get one good one and they cost an eye-watering amount 🙂
Intermods and PTFE don't get on well. Oh and H&S get upset when your cavity filter is turned into noxious fumes in the lab. In space the client gets really miffed.
Intermods and PTFE don't get on well. Oh and H&S get upset when your cavity filter is turned into noxious fumes in the lab. In space the client gets really miffed.
My work with laminates at the moment is more physical properties, but they tie very closely to the ele props.
Laminates will have anisotropic thermal expansion coefficients, strength, modulus, as well as electrical permittivity along all 3 axis of the material. Thickness characteristics will depend heavily on how compressed the board is during manu. Length and width will depend on glasscloth weight and weave design.
Surface roughness will impact hf signals, and even if the surface is smooth, the underlying glasscloth texture will be apparent to the hf signals as a permittivity roughness..
IN SPACE, NO-ONE CAN HEAR YOU "MIFF". 😉
John
Lol. But the men with coded shoulderpads are very audible (think stop that pigeon).
Last edited:
All these are covered in the IPC specs and material data sheets... On site so cant find the generic table covering most of the materials and their physical properties.
There are laminates with a smooth copper surface on both sides again to avoid the impedance discontinuities that standard copper has on the non rolled side has.... There are also as said finer weave patterns for glass to again avoid these problems, that are more noticeable on diff pairs especially if you route them on a loose weave laminate and one leg is over the glass and one leg over the resin. Thermal expansion is worse in the Z- direction for glass reinforced laminates, some exotic laminates are terrible.
What does surprise me is that some (quite a few) that are concerned about the sound of caps and resistors and other esoteric problems, yet seem to ignore the technology surrounding the MOST important component in any electrical assembly the PCB, and for audio a good quality, high Tg FR4 material is perfect, easily processed by the manufacturer, well known and studied parameters and can be used for almost anything... The use of exotic laminates is actually very low, though increasing it is still in single percentage figures I believe.
Shhh Marce, if you let them know they'll suddenly decide red resist sounds better than blue or green!
As risk of opening another can... Other than mixed signal work is there ever a justification in audio for 4 layer boards assuming a significant level of competency with layout?
As risk of opening another can... Other than mixed signal work is there ever a justification in audio for 4 layer boards assuming a significant level of competency with layout?
here we go again on teflon pcbs 🙂
holy crap you guys do not want to give up on the stuff.
Like any engineering project, you have to weigh your costs vs performance.
Using anything other than epoxy/glass is exponential in cost and certainly not exponential in performance at audio frequencies.
FR4 is your flame retarding property
Tg is your plastic state temp
The mil specs call it G10 expoxy/glass material.
If you do not run your circuits hot or your ambient is norm room, then you do not even need high Tg material
people mix up these names, not knowing what they mean, some are the same thing, just called up in different specs, mil vs ansi/ipc.
Look at a good pcb fab to get their docs of what they offer and recommend for the application.
Oh I recall saying you are a half a$$ EE if you can't lay out your own pcbs 🙂
back in the day, some of the EE's sat beside the lady running the cadnetix ecad saying move it there, route it that way. okay done run the photoplots, get check plots for review before we fab.
holy crap you guys do not want to give up on the stuff.
Like any engineering project, you have to weigh your costs vs performance.
Using anything other than epoxy/glass is exponential in cost and certainly not exponential in performance at audio frequencies.
FR4 is your flame retarding property
Tg is your plastic state temp
The mil specs call it G10 expoxy/glass material.
If you do not run your circuits hot or your ambient is norm room, then you do not even need high Tg material
people mix up these names, not knowing what they mean, some are the same thing, just called up in different specs, mil vs ansi/ipc.
Look at a good pcb fab to get their docs of what they offer and recommend for the application.
Oh I recall saying you are a half a$$ EE if you can't lay out your own pcbs 🙂
back in the day, some of the EE's sat beside the lady running the cadnetix ecad saying move it there, route it that way. okay done run the photoplots, get check plots for review before we fab.
- Status
- Not open for further replies.