#81 MOTU Traveler teardown & upgrade

As a follow-up to the recently posted MOTU Ultralite teardown, i had to grab this as soon as it popped up in my search results, because it had a price i couldn't ignore or refuse, and it had the ideal feature set for recording my band's jams and rehearsals. While the former had most of the features i needed in the 828mk2 i had been using, but in a smaller form factor, this Traveler packs more of the features i need (two extra mic preamps, while retaining the -10dBV/+4dBu line inputs with optional 6dB boost), in an intermediate form factor, as well as having a full 53dB of gain available on the mic preamps, but without weirdo custom potentiometers to wear out. Just about best of all worlds, really, especially for the paltry 85e this cost me. And as the cherry on top, it even has a separate battery input - hello, fully mobile recording sessions...

 

 

And now, time for an extended and thorough look-see inside...

One word to describe the view would be "oxymoronic" - both fully-packed, and at the same time, also barren. But even from a first glance in there, a few things strike me as odd as well as not so odd, but all regarding the electrolytics. Nevermind that a few are bloated / domed, but it's really interesting what a hodgepodge of brands in there. Sure, Meritek have been a MOTU staple for ever(?), and i've seen the odd Elnas pop up in some 828mk2's, but those Panasonics on the mic inputs? It'll be interesting to see if those are "third party", or whether they're factory-installed.

After the diode-steering of the various power inputs (barrel-socket, Firewire bus power, battery input), the main supply voltage goes through a paralleled pair of JRC NJM2397 linear regulators, interestingly enough. They seem to be set for an output of about 20V or thereabouts, so they act more rather as a voltage clamp / limiter than anything else. If the input voltage is below the set output, there's no regulation going on, save for whatever the (minimum?) voltage drop across them is. Interesting choice, but who am i to argue?

The aforementioned DC rail then feeds the four switching regulators found on board. Three of them are Linear Technology (now Analog Semiconductor) LT1767 step-down regulators, for the +3.3V and +/-5.8V supplies, while the fourth one is the already familiar Texas Instruments TPS61040 boost-converter for the phantom power.

Further down the board there's a National Semiconductor (now Texas Instruments) LM2990 negative linear regulator, cleaning up the power to the mic preamps.

And not too far from the former is a Texas Instruments TPS79601 filling the same role, but for the positive supply rail of the mic preamps.

As a convenient transition from the power regulation circuitry to the digital side, the recurring Texas Instruments TLC2933A PLL is powered by a National Semiconductor / Texas Instruments LP2950 3.3V linear regulator.

The brain of the whole operation is one familiar from the 828mk2 - an SST89V564RD microcontroller, while the SST25VF512 next to it most likely stores the various settings.

Then, the... umm... skeleton, i suppose, is an Altera Cyclone EP1C3T144 FPGA, most likely dealing with marshalling the audio data in and out of the DSP (oops, spoilers - but then again, it was visible in the birdseye view, so...), as well as the Firewire interface chip. Speaking of which, that's a Philips PDI1394L40BE link layer controller. That then connects to a "classic" Texas Instruments TSB41AB2.

The above-mentioned (as well as above-pictured) DSP - let's call it the brawn - is a Texas Instruments TMS320VC5502, and a(nother) TI TPS730 provides its 1.2V core supply.

This side of the silicon is directly and closely related to the guts of the 828mk2, which shouldn't really be a shocker at all. Why reinvent the wheel, right? The processor in there didn't have an external flash memory though, despite the feature set being quite comparable.

Moving on to the preamps - now we start to see commonalities with the "mk3" generation, in the four Texas Instruments PGA2500 digitally-controlled preamplifiers. The neighbouring ST TS922's may or may not be buffering the instrument inputs. Interesting that there's no mention of input resistance / impedance in the datasheet, but... I guess it's perhaps so high, that it's irrelevant?

The headphone output is a bit of a throwback to the Ultralite teardown, actually - an AKM AK4382A stereo DAC, an NJM4580 providing the low-pass filtering, and a TS922 as a high-current output buffer.

Four AKM AK5385A's provide the analog-to-digital conversion for the four mic/instrument inputs and the four line inputs. If you're getting a feeling of deja-vu (yet again), they're the same converters used in the 8Pre, in fact.

The digital-to-analog conversion is performed by an AKM AK4358 8-channel DAC, as seen before in the 828mk3 and Traveler mk3. Can you see a pattern developing yet?

The rest of the analog circuitry ("part one") is made up of three NJM2737's for the two line inputs on the mainboard, and NJM2115's for post-DAC filtering as well as NJM4580's for the four line outputs.

The analog daughterboard simply mirrors / clones the circuitry it sits above.

Now, with all of that out of the way, time to move on to the upgrades for this thing. Obviously and first of all, replacing all the domed electrolytics, but more likely all of them (if only for peace of mind). Then, replacing the DC input jack (5.5/2.5mm jobbie) with a Cliff FC681465, which has a spring-contact center shaft(?), that can accept both 2.1mm as well as 2.5mm plugs. Just for the sake of flexibility, really. Thirdly, umm     ... Nope, that's about it, can't really think of anything else i'd want or need. Except for putting this to work, of course.

Ended up with a handful of Rubycon MBZ's and YXF's, an older but still in-spec Elna RJ3, a couple UCC KY's and some Panasonic FR's. Barring one solder whisker bridging a newly-installed cap on the -5.8V rail (quickly disposed of, though), all is well. Apart from the different complement of ins & outs and the 176.4/192kHz capability, i will say one thing: this boots up a hell of a lot faster than the 828mk2, despite having the same processor - go figure... Either way, i really don't think this little fella's gonna be departing from my arsenal anytime soon.