|
Post by gouge on Mar 18, 2018 2:34:03 GMT -6
pretty disappointing when i hooked up my rma unit for the first time at a session and noticed that 96k has a hiss. it's so audible i had to not use the unit.
the cost of this thing is mounting. i bought it second hand and it was damaged in the post. another $100 to the tech to replace the front switches and resolder a wire internally that had come loose and everything is working as it should. except 96k....all other sample rates are good.
who fixes this kind of thing. ill give totally technical and proharmonic in Australia a ring but am not sure what other options there are.
Jim Williams maybe.
|
|
|
Post by indiehouse on Mar 18, 2018 5:38:17 GMT -6
pretty disappointing when i hooked up my rma unit for the first time at a session and noticed that 96k has a hiss. it's so audible i had to not use the unit. the cost of this thing is mounting. i bought it second hand and it was damaged in the post. another $100 to the tech to replace the front switches and resolder a wire internally that had come loose and everything is working as it should. except 96k....all other sample rates are good. who fixes this kind of thing. ill give totally technical and proharmonic in Australia a ring but am not sure what other options there are. Jim Williams maybe. Did you hit the reset button on the front? You gotta reset every time you change SR.
|
|
|
Post by gouge on Mar 18, 2018 5:42:40 GMT -6
no i didn't.
i'll give that a try this week and see how it goes. have you had a similar issue?
|
|
|
Post by indiehouse on Mar 18, 2018 5:53:39 GMT -6
no i didn't. i'll give that a try this week and see how it goes. have you had a similar issue? Yeah. You get noise if you don’t reset.
|
|
|
Post by gouge on Mar 18, 2018 6:12:42 GMT -6
ok thanks, that's a big help.
|
|
|
Post by svart on Mar 18, 2018 7:19:06 GMT -6
I dunno why RM didn't incorporate the reset in his sr switching. I did. When you hit the switch to change sr on mine, it mutes, changes sr, resets, then unmutes once it relocks to valid data.
It wasn't that hard.
|
|
|
Post by BenjaminAshlin on Mar 18, 2018 23:29:15 GMT -6
I dunno why RM didn't incorporate the reset in his sr switching. I did. When you hit the switch to change sr on mine, it mutes, changes sr, resets, then unmutes once it relocks to valid data. It wasn't that hard. I think its amazing Ross advertises PCB designing. His PCBs are cobbled together from adapter boards I'm surprised that half the RMA units even work.
|
|
|
Post by svart on Mar 19, 2018 6:39:27 GMT -6
I dunno why RM didn't incorporate the reset in his sr switching. I did. When you hit the switch to change sr on mine, it mutes, changes sr, resets, then unmutes once it relocks to valid data. It wasn't that hard. I think its amazing Ross advertises PCB designing. His PCBs are cobbled together from adapter boards I'm surprised that half the RMA units even work. I'm not sure they do.. Lots of people complaining about small things like uneven noise floors, soft hiss at one or more settings, random failures, etc, even after being "repaired". I've seen a handful of them and each one was indeed cobbled together with little adapter boards, loose wires, hot glue, and all kinds of shoddy assembly. I've even tried to repair a few for folks, but it seems like there is all kinds of ground loop issues with these to go along with all the other issues. Anyway, I always thought it was funny that he started selling his "new" versions with "4 layer PCBs for better performance" a few months after I introduced my design, which used 4 layer PCBs.. But I suppose I expected it since I sent my prototype unit around for people to check out and test, and I know it went through some RM "friendly" folks, and was disassembled at least twice while it was out..
|
|
|
Post by pope on Mar 19, 2018 9:35:13 GMT -6
But you can't really "copy" a 4-layer PCB, can you?
|
|
|
Post by svart on Mar 19, 2018 10:42:05 GMT -6
But you can't really "copy" a 4-layer PCB, can you? To some degree you can if you make some educated assumptions about the inner layers, but I also explained in my lengthy design thread that the inner layers are power planes for super low impedance supplies to the various sections.
|
|
|
Post by rocinante on Mar 22, 2018 9:57:19 GMT -6
But you can't really "copy" a 4-layer PCB, can you? To some degree you can if you make some educated assumptions about the inner layers, but I also explained in my lengthy design thread that the inner layers are power planes for super low impedance supplies to the various sections. This got interesting. Is that standard for audio Svart? Both layers? I'm guessing it's low voltage and you might be able to put it up to the light to see?
|
|
|
Post by mcirish on Mar 22, 2018 10:50:41 GMT -6
On a four layer board it would be hard to see the inner traces as they are more "floods" that cover most of the layer. You can pretty often assume the only thing happening on the inner two layers is VCC and ground.
|
|
|
Post by svart on Mar 22, 2018 12:18:46 GMT -6
To some degree you can if you make some educated assumptions about the inner layers, but I also explained in my lengthy design thread that the inner layers are power planes for super low impedance supplies to the various sections. This got interesting. Is that standard for audio Svart? Both layers? I'm guessing it's low voltage and you might be able to put it up to the light to see? I don't think it's standard for most consumer audio, but I'd bet that the pro-caliber stuff is at least 4 layers. In my day job I MUST use specialized PCB design, and some of the stuff that seems like ridiculous overkill for audio is bare-minimum for high speed work, but it helps keep things clean and tight. PCBs don't cost nearly as much as they used to, so having more than 2 layers is not prohibitively expensive anymore. I think the cost difference between a 2 layer board option and the 4 layer option I went with on my design was only a few dollars. For super high volume stuff like Behringer, that might mean hundreds of thousands of $$ over the lifetime of a product, but for something that sells a few dozen/hundreds a year, that's nothing if you're gaining performance and reliability. I'm sure RM has had to repair tons of his units due to the wiring and (IHMO) shoddy assembly. if I spend more money on a PCB that allowed me to have very little wiring, that's money I saved in the end by not having reliability issues. What I meant about layers and guessing is that anyone who's concerned about power supply cleanliness would already consider at least using ground planes, and more recently power planes. Going further and seeing that the outputs of my power supplies use wide top planes that terminate into the board with a half dozen vias would realize that there must be internal layers, as would seeing no connections to the power traces on the back side of the board for the vias that bring the power up locally to the opamps, etc. That kind of educated guessing is what I'm talking about. On mine, I used ground fill around all the parts and audio/data traces on the top and bottom. The 2nd layer was ground under all the signals and power but also some control lines and such routed away from the digital streams and analog audio traces. 3rd layer was power planes to keep all supply conductors as low impedance as possible to reduce the amount of local decoupling necessary, and to reduce parasitic inrush currents (that happen having large caps far away from the supplies).
|
|
|
Post by svart on Mar 22, 2018 12:20:05 GMT -6
On a four layer board it would be hard to see the inner traces as they are more "floods" that cover most of the layer. You can pretty often assume the only thing happening on the inner two layers is VCC and ground. Very true. I also flooded all unused area with ground for better EMI control and lower ground impedance, so that extra copper would obscure 99% of the inner layers anyway.
|
|
|
Post by rocinante on Mar 26, 2018 13:29:59 GMT -6
Awesome. Thanks.
|
|
|
Post by pope on Mar 26, 2018 15:07:22 GMT -6
Wouldn't make sense to have the 2 outer layers as ground planes for better shielding?
|
|
|
Post by jdc on Mar 27, 2018 7:44:01 GMT -6
i have both the rm pcm4222 and svart's adc/dac. i haven't used anything but svart's box for critical listening since comparing it to everything else i had (bla sig modded 002, ross martin, motu 16a). i've been buying gear and building up a studio regularly for the last 5 years and i've noticed that purchases fall into two categories: the things you buy once that get used every day, and then everything else that leaves you chasing. svart's box was a piece of the chain i didn't have to think about anymore because i felt i had solved it.
|
|
|
Post by svart on Mar 27, 2018 9:45:53 GMT -6
Wouldn't make sense to have the 2 outer layers as ground planes for better shielding? Not necessarily. it depends on what kind of noise ingress you're trying to shield. Since I didn't plan on routing digital traces close to the analog traces, I didn't foresee needing buried analog traces. Using inner layers also comes with issues like needing to use vias, and coupling noise from the outer layers into the analog traces since "ground" is also a current carrying conductor. Digital hash has a bad habit of creating current noise on ground, and it's more advantageous to just route the digital further away from the analog sections, than simply burying the analog between layers but still running below the digital circuits. I was also more concerned with ground loops than internal noises, and made use of of the chassis as a pseudo "ground bar" in the design, so the chassis is also integral in the shielding of the unit.
|
|