Front panel's pretty fully packed - look at all the buttons and 3mm LED's! Not all just for show, though. Each of the four mic / instrument inputs has an input selection switch, a pad switch, a high-pass (aka low-cut) switch, and an individual phantom-power switch.
Quite busy around the back side as well. The four line inputs reside here, as well as, surprisingly enough, inserts for all eight analog inputs (the mic/instrument ones on the front, and the line inputs on the back), and all the digital ins & outs.
The format and the feature set remind me, to some extent, of the MOTU Traveler units (sans the stand-alone configurable mixer functionality), but with added analog inserts. Also, i may very well be wrong, but i can't seem to recall any other Firewire *and* USB connectable interfaces, outside the MOTU Hybrid lineup. On a brief research, it would appear these units require a power cycle to switch between the two connectivities, but anyway... I didn't exactly buy this to use as a "daily driver".
The eBay listing said it doesn't even power on, but the seller couldn't tell whether it was a power supply issue (external 12V 2A power brick), or within the unit itself. For the record, i did indeed stick to the mantra of "Don't turn it on, take it apaaaah't!", but i did venture plugging in the power brick. At least unloaded, puts out 12.03V, just as it should. Whether it can still maintain that under load, is a totally different story (as a local friend whose Scarlett 18i8 wall-wart i recapped last month, can attest).
That being said though, there's something rattling inside the casing. Could that perhaps be a clue? Off comes the top - drumroll, please...
Well, you don't see a yellow-soldermask board all that often. I think the only other time i have, in recent times, was in some internal photos of a Behringer amplifier, on a forum. Come to think of it, there was the odd PCI soundcard back in the day with orange(ish) soldermask... Ah, the "good old days" - simple(r) times...
Buuuut back to the task at hand. Spoiler alert - don't look on the left side of the board!
Glossing over the arguably... interesting area of the board, the main controller is an Archwave DM1500, and as one might expect, some DRAM (Micron MT48LC4M16A2 64Mbit / 8Mbyte SDRAM) and firmware flash (Macronix MX29LV320EBTI-70G 32Mbit / 4Mbyte) in the vicinity. Also, a near-classic TI TSB41AB1 is handling the physical / electrical interfacing to the Firewire connection. And last but not least, "below" all these is a small TLV0832 analog-to-digital converter - pretty sure this reads the "Mix Bal" and "Main Vol" potentiometers on the front panel and passes the data on to the brains of the operation.
Further on, we have the converter section, all Cirrus Logic equipped. The bigger chip is a CS42435 codec, containing four analog-to-digital converters and eight digital-to-analog converters. The two smaller ones are a pair of CS5340 A-D converters. Can't quite tell which are in charge of the front inputs and which handle the rear ones, but either way, one can be sure they haven't "interleaved" them.
Despite Behringer having a reputation for (nearly) bottom-of-the-barrel equipment, they haven't cheaped out too-too much, at least in this case. All the silicon looks to be name-brand, even the opamps. Most of them are SOIC8-packaged 4580's, but going by the font and the -R suffix, i'm reasonably sure they're Rohm BA4580RF's.
And now, we get to the mic preamps. Same opamps as mentioned above, and a somewhat familiar sight yet again - a pair of 2SA1268GR / KTA1268GR PNP transistors. Between these and the four slightly fatter 8mm diameter electrolytics in the area, this "smells" to me a lot like the "XMAX" preamps in the Presonus Audiobox 1818VSL from a while ago - in other words, a pretty classic circuit.
The board also contains four SIP9-packaged NJM4580L's, most likely to drive the two headphone outputs, given their proximity to the connectors for the cables going to that area. Once again, nothing space-age, but no no-name parts employed. On the other hand, though...
Umm... Yeah, i think i finally found the cause of the rattle mentioned near the beginning. Whoopsieee...
Surprise-surprise, power supply did a boo-boo. And, "of course", took the main PWM controller down with it.
So then, after the DC input jack, the two power lines go through a rectifier bridge made up of four diodes, gets filtered with the bigger electrolytic, then gets OR'd with two diodes - the other one, i suspect, being the power supply coming from the Firewire connection (when used / available). That then gets filtered by the destroyed capacitor (which used to be a 47uF 25V Decon SA).
The sadly deceased PWM controller, after some digging (due to half the marking being illegible due to the crater in it), turned out to be a LinearTech LTC1871. That, in turn, drives a FairchildSemi FDS3672 N-channel MOSFET, which then drives the flyback transformer. This creates the +/- analog supply rails for the opamps, as well as the phantom-power rail.
On the other side of the blown electrolytic is a Monolithic Power Systems MP1591DS step-down converter, which most likely handles the main digital power rail - my best guess would be 5V for the converters and likely LED driving (on the front panel), which then gets regulated down to 3.3v by the SOT223-packaged BCD Semiconductor AZ1117.
[To be continued...]
The FCA1616 was said to be an okay interface if not for its tendency to, well, explode.
ReplyDeleteI have one in the rack that never gets used because of this.
Any insights into what defect causes this?
Not sure i can say. Well, that little exploded electrolytic is in a position that, i would think, is subject to a considerable amount of ripple. I am quite surprised about the PWM chip giving up the ghost, though, despite the "heavy lifting" circuitry still being fine.
Delete"By the way", i finally got around to replacing that capacitor (Rubycon ZLH 220uF/25V) and the PWM chip, a couple weeks back, and it's back to life.
Was researching on a similar fault for a friend. Haven't seen his device myself yet. Here is the schematic. IMHO the PSU is an extremely greedy design. Probably the analog rails are quite noisy. Also I can not imagine that without any feedback from the secondary, the +15V and -15V would have symmetrical and precise voltage. Wondering whether this topology that basically has no regulation (and only 147uF total capacitance on the secondary side for each rail) is even capable of providing enough current for all those dozens of opamps under load.
ReplyDeleteAs to why C30 commits suicide in these devices, I'd guess it's because C30 is the only "bulk" capacitor on the primary side and it simply can not handle so much current.
Would be interesting to hear your thoughts.
https://drive.google.com/file/d/17W-MV_yg0dkG0bn6FMaWxGqrGunVHH1H/view
"Greedy", in what way?
DeleteAnd how do you figure there's no feedback? I see a node / tag called "FB" going into the R94/R93 divider, that's coming from the +15V rail.
That's pretty much what i figured as well - a teeny-tiny little 47uF cap (and not even a low-ESR one), on the input of not one, but TWO switching converters, is indeed a recipe for disaster.
Big thanks for the linked service manual, though! I didn't even imagine anything like that would be available anywhere, but nice find!
Yeah, I was as surprised as you are at the luck of finding the schematic on some Russian forum. Greedy for a few reasons. First, because they obviously skimped on capacitors. Second, because individual rails are not regulated. You are right as to +15V rail though: I didn't notice the feedback path, my bad. Well, at least something. The load on the +15V and -15V rails is asymmetrical though (considering that the +5V rails are derived from the +15V rail), so in the absence of regulation on the -15V rail, there quite obviously should be voltage asymmetry. Linear regulators for the analog rails do not exist as well, which, combined with the skimping on capacitors, means there should be quite a bit of noise. Then all that noise goes into the rails feeding not just opamps but also discrete transistor stages in the preamps. The preamp incorporates a cascode topology which should be able to provide an improved PSRR, so at least the noise injection should be mitigated to some degree.
DeleteWell, i'd call that the complete opposite - "cheap" (rather than greedy). But then again, "a penny saved is a penny earned"... Especially when you make thousands (if not millions) of items.
DeleteWell, without actual measurements, you can't really tell / know just HOW noisy or asymmetrical the power rails are. And even so, since those SOT-223 packaged linear regulators are only good for up to 1A (best case scenario), and the rectifier diodes are rated for 1A in the first place... Yeah, i wouldn't worry too much.
If you're really curious, terminate a mic input with a 150 ohm resistor between pins 2 and 3 (it's a standard for specifying mic preamp noise), and see what dBFS noise level you get at various gain settings. Without that termination, things WILL be noisier.
The preamp design seems very uncommon BTW. I've been trying to figure it out and spent a few hours looking at various topologies. Haven't found a single one that resembles this, incorporating in the first stage exclusively PNP transistors in this particular cascode configuration (or what I take for cascode), plus the negative feedback going into the upper cascode transistor (which seems to be a very uncommon practice on its own). That is until I found this patent.
Deletehttps://patentimages.storage.googleapis.com/52/76/46/8792fb86ea4215/GB2242089A.pdf
It looks remarkably similar, except it has a current source instead of just a resistor, and the negative feedback in the patent goes into the emitter of the cascode transistor instead of into the base as in the Behringer. All this makes me very curious, unfortunately I'm a beginner when it comes to amplifiers, so can't really appreciate the differences in the topology. Would be interesting to hear your opinion.
It sort of reminds me of the Audient preamp i deciphered in an iD22 interface (teardown article coming soon). It's kinda-sorta similar to this, but this looks like a cut-down version of that.
DeleteUnfortunately though, this level of complex(ish) analog design is a bit above my pay-grade as well :)
Interesting. The two guys who founded Audient were previously engineers for Soundcraft, which filed the aforementioned patent. So I guess it could make sense. We'll see :)
ReplyDeleteWell, as i've mentioned in another comment here - a tiny little 47uF, probably not even from a low-ESR range, not to mention the questionable provenance, on the input of a flyback converter... A disaster waiting to happen, really.
ReplyDeleteThe inductive "dump" would make sense though, now that you mention it - good thinking!
My friend bought one of those..but the internal i/o's won't work. I opened the fca1616 and saw, that there is no trafo (tr1). Can you Tell me the Type of it, for replacement?
ReplyDeleteUnless you're willing to wind your own, i'd be totally shocked if you managed to find one on the retail market (even from big component distributors) that would have 1) ALL the necessary windings, 2) with the correct pinout and 3) in a suitable footprint.
DeleteThen again, knowing the input and output voltages, figuring out the turns ratio for a flyback "transformer" (technically it's a coupled inductor) shouldn't be too difficult. The headache's the practical / physical side of things.
If you don't have the trafo, and you don't mind missing the phantom power, you can always wire in a DD1912PA module which takes 12V DC and gives you +/- 15V. It works in my application where the unit was fixed before and the pads for IC7 where messed up. I mainy use this interface for testing purposes, but so far it works like a charm. This is also coverd in a Youtube video, where the guy wires in an external PSU. For me the phantom power can be missed and DD1912PA is much much smaller and can live inside the unit.
DeleteWith one of those supplies, if you're willing to mod the thing, you might even be able to hang a voltage multiplier off the positive output rail, and derive the phantom power voltage that way. Especially since that doesn't need that much current, to begin with.
DeleteHi, thank you :).
ReplyDeleteIf someone takes the trouble to determine the electrical resistance, the wire diameter and the material, you can do that with
R=rho*l/A
calculate. You would only have to have the diameter of the winding bodies. In the device the part is missing but completely, I can not measure it
That's a big "maybe", especially with low numbers of turns. Either way, i'm afraid i can't help you, since i've sold this unit off sometime last year.
DeleteWhere can iget this ic1817 lanear
ReplyDeleteI'm assuming you mean the LTC1817 PWM controller? If memory serves, i got mine from eBay, from a Polish component distributor.
DeleteI always do this repair, always changing the capacitor putting a 47uF×100V. And you can change the LTC1871 too. The cause of the burn is the capacitor, once I changed the LTC and forgot to put another capacitor... it blew right away!
DeleteI would argue the 100V rating is unnecessary / overkill. A good-quality 35V one, with as much capacity (uF) as fits into that footprint, and preferably low-ESR, is all that should be needed.
DeleteI've sold this one off almost a year ago, and i've yet to hear back from the buyer, so... "No news is good news", i guess.
Haha indeed...
ReplyDeleteSo, summarizing for trying to prevent the FCA1616 to stop working due to power regulation failure:
ReplyDelete- Rreplace periodically C30: 47µF/25V -> 47µF/35V low-ESR
What else?
My unit is degrading week after week with pop/cracks and USB disconnections. Reconnecting 12V supply heals it for a while.
Thanks!
José
You'll want to fit as big a value capacitor as you can fit in that footprint (25v minimum, and 100-220-330-470uF), and definitely from a good brand low-esr series. Although one must wonder if that's the only cause of your issues; it might not be.
DeleteThank you!
DeleteThis seems to be the most appropriate I can buy easily:
1500uF
35V
20mOhm
2880mA
12.5x35mm
0.54Eur
Do you think it is enough low-esr?
Thanks once again
Brand and series?
Delete12.5mm ones usually have a 5mm pin pitch; is that the same as on the board?
No brand unfortunately. It is from a local reputable seller though. Panasonic are only a few models available, none that fits. Mouser wants 20Eur of shipping... But there there is a Panasonic one: EEU-FR1E821B
DeleteYes, it is 5mm pitch (like the panasonic). It is bigger than the original (6mm total diameter).
Does it need to be mounted flush with the board, or can it be dangling a bit in the air?
I don't trust the specs of no-name brands. That Panasonic 820uF 25v part is much more trustable, in my book. You'll want it as close to flat on the board as you can get it.
DeleteIt seems my board (PCB213121REVD_01) already came with a better capacitor than the ones described above: CapXon 47uF 35V KF 105ºC. And it is the only one from that brand. All others are KSD.
ReplyDeleteSo, it is a very long shot to try to fix this unit by replacing that capacitor.
Will try a different power supply brick.
Thanks again.
José
"Better" in the sense of "not exploded (yet)"?
DeleteCapXon is thoroughly in the "guaranteed to fail, sooner rather than later" category. I've even seen them fail as DC-blocking capacitors, on the outputs of a dbx digital crossover(!!!), nevermind all the dozens of dead ones in power supply filtering duties in countless other devices.
eheh... I meant better than above, since mentions above are about not being special in terms of low-ESR. I guess if the one on my board would not be a low-ESR, it would be the same brand as all others in there.
DeleteTo know if C30 is the culprit or not, maybe I could replace it by a common one (no brand, no low-ESR) just to check if the existing one is already degraded and the culprit of the pops and disconnects. Then I would buy the Panasonic one as soon as my shopping list on Mouser grows above free shipping threshold. Or would this experiment jeopardize the LTC1871?
Hi everyone, I changed de the Cap and the IC and the leds turn on again, but no sound in analog inputs or outputs, only sound if I connect optical. any sugests?
ReplyDeleteSounds like something might still be unwell on the analog supply side. Are D28/29/30 ok? And the capacitors connected to those?
Deletehey, I own one of those units. mine still works but its good that I can do some pre-emptive cap replacing so as to not bring down other components that could be effected. I also recently bought the FCA610 (not working) - it powers on and registers in the pc via usb but the analog side of things is dead. I did find a puffy 47uf cap (bulging top) so I'm thinking its the culprit. Unfortunately I haven't been able to find any schematics for the unit so I'm thinking its most likely to be pretty close to the FCA1616. if anybody knows of the schematics for this, it would be helpful and as well for the FCA1616 - I had read someone has that floating around. thx in advance
DeleteThe culprit on its own, that's a big "maybe" - i haven't seen inside an FCA610 to know how the power supply section is structured. But if all the analog ins & outs are dead, the one thing those all have in common is the power supply, so that's where you'll wanna start looking indeed.
DeleteThe Google Drive link in one of the comments above, containing the FCA1616 service manual seems to still be up, and it wouldn't be shocking if the FCA610 shared much of its structure and guts with the FCA1616. Why spend time and money "reinventing the wheel"?
https://drive.google.com/file/d/17W-MV_yg0dkG0bn6FMaWxGqrGunVHH1H/view
Hi,
ReplyDeleteI am sorry to ask this here but I dont know where else to ask.
There is Phonic 808 Universal that also use the Archwave DM1500. I read from a forum/group that many of them broke in a vely similar way: can be turned on but not detected in Windows.
The inside looks like this:
https://www.imagebam.com/gallery/356a66c8d086cd4acea6fbeb40578107/
Maybe the firmware bricked?
If this is really the case, what options do I have?
I dont see firmware updater in their download center:
https://www.phonic.com/index.php/download-center/category/21-firefly-interfaces
I doubt any "firmware update" would do anything, if the device doesn't get detected by the computer (properly). Your best bet might be to flash the firmware chip itself (after desoldering it from the board), with the contents of the firmware chip from a working unit. That is, assuming it's not some quirky power issue, or interface chip issue...
Deleteyah, it does seem very close to the 1616, thx for the link, i'll start looking at the chips to see if they are the same as the 1616. Both are nice units as far as features, but behinger is noted at being cheap on parts that have half-lives, ie electrolitics caps. and thier costs are low enough not to get them repaired - designed obsolesents at its best.
ReplyDeleteI have of these, it gets, and sends output from my windows pc through usb but none of the inputs "hear" whats plugged in(guitar, etc. With a 6mm jack), I know it's not a software issue because I can't hear anything even with the no latency monitor outputs. Any idea what it can be?
ReplyDeleteSo you ARE hearing audio from your computer, through the outputs, but not receiving any from the inputs?
Deleteno I'm not hearing anything, it just shows me that it's getting sound from my pc, but I can't hear anything. I'm using a pioneer sa-055 as an amplifier and RS brand passive speakers. I don't know what I'm doing wrong because the cables are correct, the converters are correct. Also, it never shows me any input from my guitar on any input and I can't hear it from the bypasses output either.
DeleteI guess it's not completely impossible the power supply feeding the analog circuitry has failed.
DeleteIt's been a while since i had one of these opened up on my bench, but the input voltage seems to split off into the supply for the digital circuitry (which obviously works, if/since your computer still detects the device), and the supply for the analog side. As you can see in the photos here, and in some of the other comments, that side seems most prone to failure.
Pop the cover off yours and see if there's anything visibly toasted?
hi, following your instructions and i was able to fix it and it worked for an hour before dying again. to be specific, my fca 1616 also had a faulty ltc1871 chip and i ordered it, had a professional replace it and it worked, it was fine, both input and output. except it had a hissing sound coming from the capacitor next to the chip like in your post, and then ''boom'', it exploded. I think I read your article quite carefully with my basic knowledge, I brought it to the technician to replace a 25v 220uf capacitor and then it only displayed the led signal when I connected the usb to the computer, all the signal indicators were on the led and there was no sound at the output, I determined that it had the correct signal from the computer, and was somehow broken in the sound transmission and reception, I hope you can tell me the possibility of something being "broken" in this case? thank you
ReplyDeleteWell, there's no guarantee that the blowing of that capacitor hasn't re-damaged the LTC1871, the switching MOSFET, or any of the diodes on the secondary side of that power supply, or voltage regulators downstream of that. The digital circuitry (that also connects to the computer) is powered from a separate supply.
Delete