I put up a web site article at line #65784 as it is about the various FR4 and others to compare... all the pro-con are there.
As for what JC uses... he said here that the pcb choice is also based on listening evaluations.
-RNM
Last edited:
Scott
I have become allergic to esoteric’s notion of exclusive in heritage of secret "Ancient Wisdoms".
They can never provide a credible traceability back to the named origins, (which happen to be lost or never existed in written form)
George
Again this is based on higher frequency design and not really relevant to the frequencies found in audio. It sais the same as I am saying as do many many articles, for low frequency work <2GHz and in normal operating environments FR4 provides the best solution for manufacturability, signal integrity, assembly, cost etc. The parameters are well known and acceptable for all but the >2-4GHz or extra special designs.
A bit of audio equipment in a house is not a taxing problem from an engineering side, benign temp range especially when operational, low frequency range including the digital side....
Again it is only a small fraction of the information out there regarding ALL aspects of PCB design, materials, processes etc.
Here's a good start and I am sure you must already know use their stuff, just quoting the standards for PCB manufacture and assembly should be a base for anyone procuring PCBs... And here for anyone's interest is my basic PCB procurement document, that covers most aspects of what you want to control and the relevant IPC spec that it will be manufactured to....
A bit of audio equipment in a house is not a taxing problem from an engineering side, benign temp range especially when operational, low frequency range including the digital side....
Again it is only a small fraction of the information out there regarding ALL aspects of PCB design, materials, processes etc.
Here's a good start and I am sure you must already know use their stuff, just quoting the standards for PCB manufacture and assembly should be a base for anyone procuring PCBs... And here for anyone's interest is my basic PCB procurement document, that covers most aspects of what you want to control and the relevant IPC spec that it will be manufactured to....
Ah a fun time, can tell you many tales of woe regarding the quality of FR4 over those years, early 90s and before blow out from the PTH holes was a problem and a lot of materials or processes were not as refined as they are now.....
2003 the move was well under way to better quality with some high volume Japanese companies moving to lead free before the 2006 deadline and encountering many problems along the way...
There was also a glut of 130 deg C FR4 laminates and pre-pregs around, perfectly adequate for consumer/industrial designs with leaded solder, not good for lead free process, after getting stung with processing problems and high failure rates (via cracking) around 2006 when moving fully to lead free we went to 170 deg C TG FR4 and the problems stopped.
Materials then though are not relevant now, both they and the processes used have developed and improved greatly.....
At DDR3 bus speeds.....DDR5 on the graphics card possibly
FR4 is not the limit here distance is... SOC that's why that's coming along.
DDR memory appears on many designs these days, so most complex FPGA design are considered high speed (digital not RF) and require some impedance control... and skew control on all the memory lines that's why you get those pretty squiggles on your PC motherboard.
Why don't PC motherboards have DDR5 memory???
No idea. But, as an other example, TNT frequencies are up to 3 000 MHz.
Last edited:
400–1066 MHz clock speeds on board, then you have the harmonics. That's why DDR3 critical signals are LVDS, and for motherboards the whole concept including the south bridge devices etc. have been designed with the whole layout taken into consideration, including a 'war and peace' size layout guide... Also if you look carefully to achieve this with minimal layers for cost, sometimes as low as 6! but 8 is common. For better reliability and the best in signal integrity 12-14 layers with many grounds and close coupled power pairs (for distributed planar capacitance) is best, and tighter skew figures.
Interesting to look at the CTE figures in relation to Tg (glass transmission temp.) the reason I mention it quite often is if you are using lower temp laminates (cheap) with hot components and plated through holes you can stress the materials sometimes to intermittent or total failure of some through holes due to barrel cracking, caused by elevated Z axis expansion due to localised thermal stressing.
I'm going to the fridge for a beer now!
http://www.prototron.com/documents/materials/FR408.pdf
I'm going to the fridge for a beer now!
http://www.prototron.com/documents/materials/FR408.pdf
Last edited:
Consistency is a big part of the process. One of the "measures" I look for is that a particular recording always sounds the same - a big giveaway with typical audiophile gear is that if you try 3 different systems with a known recording, then you get WeirdA, WeirdB and WeirdC - each is adding a whole mix of "effects" which the owner has chosen, intentionally or otherwise, to flavour what he hears. When I start cleaning up a setup a strong indicator of progress is that its intrinsic, raw "weirdness" steadily disappears, until I can only hear the "sound" of the particular recording.
As with video. I always pay special attention to news style footage of normal outdoor scenes, in the suburbs and in the bush - this has a certain colour balance that one knows by heart. If this is subjectively a perfect match then it doesn't matter what games the producer plays with the balance in some show, it's quite easy to pick that fiddling has been done, and that's acceptable in the context of that broadcast.
Marce, what software do you favor for layout? I suspect it will be out of my price range for the moment, but I've been concluding with some reluctance that I need to do at least some of my own layouts.Well another
Brad
I've use mainly Cadstar, since it was called Redboard in 1985ish, I have a few of years experience with Allegro and play with Altium when I have time... but Cadstar is my chosen one simply because I have used it so long and have been trained (often directly by Zuken) on using it and the SIV and PIA add-ons, I prefer Inventor or SolidWorks for 3D assembly, I don't like any of the PCB package add-ons.
- Status
- Not open for further replies.