The joy of a repairman rewarded by their work is infectious. As an aspiring electronics repairman who's stuck in a dead-end job, I hope to one day be half as good as you Tony. Magical work.
@@tony359 No, thank you Tony. Your channel has been an invaluable resource for amateurs like me. Also been entertaining. Your discipline is enviable. I would've thrown in the towel after the USB port lmao.
It could have been. I hope it was! But the whole device was literally covered in grime when I got it so... well, let's be optimistic and say it was TPU :) Thanks for watching!
Great job with the repair, and even a greater job with the video editing, multimeter/oscilloscope shots, schematics and overall description of the whole process. Always a pleasure and very informative.
New QFN chips that have likely been sitting around for 10+ years tend to not stick unless you tin the pads under the chip first. So add flux and solder and tin the pads then when you hot air the chip it will automatically stick and all pads will be soldered. Also, you made the right call by wicking away most of the solder on the middle ground pad. Too much solder can squeeze out and create shorts on the outer pads which would be bad, most likely resulting in having to remove the chip and start over.
@@alelondon23 Louis is actually not that good. Yeah to the below-average noob he looks like a god but vs a pro electronic engineer he's pretty average. I've publicly documented smd soldering repairs 10X harder than anything he's done. Look them up you might learn something lol
@@tony359 imho (and just for reference I'm a test engineer for a PCB assembly company so a lot of times I have to deal with soldering defects lol :P), you could have even removed all of the solder and still have gotten away with it. It's mostly used as a heatsink/ground pad and even the small amount of solder left there would have still made a good connection and besides it's a sound card, fairly low USB data transfer rate so the chip won't get hot. The most important thing with soldering QFNs is for them to lie absolutely flat on the PCB surface. Even minor level imbalances are annoying af to deal with. You should also have removed the capacitor from the side so that you could have better accessibility. It's a bulk decoupling capacitor, even if you ripped the pads you could still hack it back together and it would have worked :P
Your thought process troubleshooting was so great to learn from, thank you for keeping it in the video. And what a satisfying repair! I hope Focusrite see it and realise how important it is to share schematics and firmware versions for their community.
28:11 I had a moment like that when i replaced the HDMI connector on a PS4. Some of the pads were broken off. I spent hours with tiny strands of copper wire to repair the traces and it looked ugly. I was jumping up and down in excitement, shouting: "YES YES YES! It works! YES!" when it finally worked. I couldn't believe it. It was such a relief and I was so happy.
That 486dx2/66 casually laying on the bench. Anyway, great repair. Random Italian "cursing" when something unexpected happen is just icing on the cake. :)
ahah - but I did not curse! :D That 486 was passing by, waiting for its motherboard to be fixed (coming soon when I get new tweezers!) Thanks for watching!
Great diagnostic work as per usual. You might want to give solderpaste a go instead of reapplying solder to the pads. That way you can skip soldering pin for pin. Clean the pads, apply a small amount of solderpaste, put the chip in place and blast it with hot air.
@@tony359after some practice its alot easier. Just get some of the cheapest chips you can find in the package you want to practice and some practice pcb's. Recently got myself a bunch of 4 layer pcb's with big internal ground planes made just to practice soldering VQFN packages for an upcoming project.
Ciao Tony. I was the service manager for Technics musical instruments for the entire life cycle of Technics. I love the way you explain what you are doing. It's not easy for one to make a living repairing digital circuitry. Customers have no idea that we work at microscopic level. I am an Italian national and your English is really brilliant. I heard the Italian long eeee sound where your pronunciation of the word ' this ' (singular) sounds exactly like ' these ' (plural). If you listen to an English speaker you will hear the word ' this ' sound a very short I sound with a long S thîssss and ' these' sounds like theeez.
Thanks for your comment, for watching and for your kind words! My brain has a limited processing power, if I focus I think I can say "this" and "these" properly but if 'm doing three more things... well the language centre goes into default mode and then my Italian re-surfaces a bit! :) At least I don't say "theess-e" :) (well, I hope I don't at least!)
yes. you can clearly hear the italian accent. but (at least to me) it's rather charming (at least much better than german accent in english (i am german)) 😁
Muchas gracias por tu aporte. En mi caso es una iTrack Solo que de la nada la pc dejó de reconocer la Interfaz La solución que me funcionó fue usar la estación de calor en el circuito usb3343 para que se vuelvan a soldar los pads, no fue necesario cambiarlo Dejo aquí el comentario por si a alguien le sirve :)
Thank you - someone else also mentioned that reflowing the USB3343 was all it was needed. Who knows, maybe it would have worked in my case too! Thanks for commenting!
You got a nice pick there... I always used MOTU soundcards, but for the money the Focusrite Scarlett was a formidable little soundcard especially for it's price point!
This was great! I'm secretly happy you didn't get this to work on the first attempt...proving you are human like the rest of us lol! I'm often working on vintage audio gear from the 1970s... really have to be careful with those solder pads as the bond between the board and the pad itself will fail even if you look at them funny. It was very rewarding to see this device come back to life. I really liked the segment where you pulled the firmware chip to validate. Fantastic as always!
thank you for your kind words! Yes, it was that good balance between "easy" and "challenging"! Just the socket would have been boring :) Thanks for watching!
Very good idea! I actually removed some from the ground pad as I thought it would be too much? I have a feeling "more is better", the excess would squeeze out when you press the IC! Thanks for watching!
@@tony359 More isn't better, not even close, less is good, but the trouble is, this IC probably spent years nobody quite knows where, its entire bottom is likely oxidised and may not be eager to accept solder before you re-tin it. I think that was why you had unsoldered pins. As such this is indeed a good call. Maybe your flux is also not quite angry enough to deal with that without a little extra encouragement. Serious manufacturers are highly reluctant to use ICs whose manufacture date is more than about 2 years old for this very reason, this is how they end up cheaply on the Huaqiangbei market.
Very difficult little chip of the devil there to solder indeed Tony... But glad you eventually persuaded it to stay at its seat completely, but the most important thing here is the correct troubleshooting method you used to discover that tiny faulty IC. I was a bit scared with the firmware situation cause you did not say, do you run the card with 214 firmware or reverted back to the original that the little ROM had inside? Also the black knobs seem to have that velvet feeling coating which eventually turns to a gross glue on plastics and looks dirty without being. Cheers from Greece, keep up, best wishes and better year for all of us than that shitty 2023! Jim.
I did revert to the 2i2 firmware, it just didn't make it in the final cut! :) Yes, that "grime" could have been that "soft rubber" which disintegrates over time. But I am not 100% sure, the whole thing was honestly covered in grime when I got it! :D Thanks for watching!
Loved how you described in detail, what tool you use for what purpose in your hunt for the issue. Made it even more fun for me to follow along. Thank you!
I have the 2i4, and it's developed a slight static in the outputs when I change the volume knob. I've been too intimidated to try cleaning it, but watching you do this repair has shamed me into trying it. :)
Thos pots are more or less sealed. You can try with some cleaner but make sure you use the "lubricated" one, for trimpots. Failing that, you could replace it. Good luck and thanks for watching!
Great work! Replacing this kind of devices has no monetary value (a working unit should go for around 30-40 USD second hand) but fixing any devices instead of buying a new device is always good for the environment.
Good investigation, analysis, and fix! This reminds me of what my team and I were trying to do when we were setting up a streaming system during 2020/COVID. While we were pretty good at computers, we were still trying to figure out how to spell "audio-visual", and all the unfamiliar hardware bits and pieces to bring sound and picture together for a decent stream. And just like you did here, it all eventually came together and it just worked! Great job, Tony. I really enjoy your presentations - keep them coming 👍😁
Nice work! Those J-lead style packages are such a hassle to solder, even for larger ICs, and because of the angle it's not so obvious when you have a bad joint.
Just a minor correction: that isn’t a J-lead package. J-lead means packages that really have pins that come out the side and then fold down, with a curved end folded under the chip. The most common J-lead package is the PLCC (though not the version used for LEDs).
I understand how you feel doing this type of work. I just did my first surface mount ic like this and if I were being paid they would have fired me after the second hour. 😂
Thank you, I appreciate the feedback! I'm always trying to find a balance between an enjoyable length and good content. Your feedback is important! Cheers!
I need to replace one of these on my Macbook pro and couldnt find a video for a quick tutorial to make sure was doing it best way possible, amazing video thank you! I never thought about soundcards using same USB IC controller.
This USB port can be changed from ppl, that solder for one day. that was an easy fix. Sorry for the lost pad - but it was toooo much heat for that little guy.
The third gen focusrite I tested had a big peak in noise at 90KHz. A problem I noticed on a lot of other interfaces by other brands. While you can not hear it, it does effect the next opamp in the chain.
That's the shaped noise from the DAC. The CS4272 DAC is a relatively old and inexpensive design with like 1 or 2 bits only that requires fairly high-order noise shaping (4th or 5th-order I think) for it to achieve its rated performance, resulting in substantial amounts of ultrasonic noise that requires dedicated external filtering in addition to what's built into the chip... which in turn inexpensive interfaces will commonly skimp on, and at least the 2i4 is no exception. The "big peak at 90 kHz" isn't even an actual peak, that's just the effect of ADC filtering kicking in when recording at 192 kHz. Things merrily keep rising beyond that point for a good while longer. Seems to be about a 3rd-order slope, so even the circuitry suggested by the datasheet would not be entirely adequate if you are serious about your hi-res.
@@PileOfEmptyTapes The codec may be an old design, but it is also not understood. Otherwise someone would place an analog low pass filter on a higher end interface. I ran a test where I mixed a guitar with a 32KHz sine into a distortion box. The poor opamp got so saturated reproducing the 32KHz tone that it undistorted the guitar. It is my understanding that a codec works by sampling at a much higher rate and lower bit depth into an adder accumulator. (CIC filter) Then you carefully select a series flat-top and or half-band filters to push the noise out of band as much as you can while reducing the sampling rate. I guess you are saying that the noise is the result of inadequate digital filtering for an interface rated for 192 KHz and a lack of any analog filtering.
@@moddaudio "The poor opamp got so saturated reproducing the 32KHz tone that it undistorted the guitar." Congratulations, you have proven that high-frequency AC bias helps a baseband signal overcome a strong nonlinearity. Messrs. von Braunmühl and Weber really were onto something, eh? ;) The importance of this effect for audio reproduction cannot be overstated. It helped tape overcome its scratchy lo-fi beginnings and turn into an actual hi-fi medium, and without it things like delta-sigma A/D and D/A converters (which are vastly dominating audio applications) or Class D audio amplifiers (which are using a switch-mode output stage and PWM) would not exist. The CS4272/CS4392 presumably runs a dual bit modulator (3 levels or ~1.585 bits) at 128fs (12.288 MHz at 48 kHz in). Even including 21 dB of oversampling gain, it would only achieve a dynamic range of about 32 dB when applying flat (white-noise) dithering, absolutely pitiful performance. These kinds of modulators need substantial amounts of noise shaping to shift most of this noise outside of the audio bandwidth, which results in a noise floor rising with frequency - so you get less where it matters and more above. The more dynamic range you are trying to squeeze out of a given modulator and sample rate by increasing modulator order, the steeper the rise is. While there is some internal analog (switched-capacitor) filtering in the DAC to address ultrasonic noise output, it is unlikely to be comprehensive in a price-conscious DAC design in particular. The maximum modulator order I've seen in a DAC has been 5th order, and when designers can get away with 4th or 3rd order while still achieving the desired performance they will happily do so, just because it makes them easier to manage. The maximum on the ADC side has been 7th order, which is really extreme but not such a big deal there as digital filtering can get rid of the noise during decimation. Even getting such a beast mathematically stable would have been quite the accomplishment (it used to be a real problem for the design of higher-order modulators). At least running a higher-order modulator means that increasing the oversampling ratio (i.e. its operating frequency) gives major gains - at 4th order or a 24 dB/octave slope, doubling it turns a converter not even quite making it to CD quality spec into a midrange audio interface affair, assuming the analog side can keep up. A lot of designs made the jump from 64fs to 128fs in about 1995-98. I realize this is a real rabbit hole into the nitty-gritty of digital audio. Understanding dithering with or without noise shaping is not trivial. Fun fact, shaped dither was instrumental in the CD revolution itself. Remember that Philips only had a 14-bit DAC at their disposal (the TDA1540)? Using a digital filter IC to run it at 4X oversampling gave them another 6 dB, but that would still only have been 15 bits effective. So the filter would apply some noise-shaped dithering as well, which pushed performance levels in the audio band to about 16-bit level as desired. Mind you, the filter created some new problems - a 0.2 dB periodic passband ripple spec in an FIR filter is not pretty, for one.
Very nice. Do you have any videos on that whole flux/soldering/cleanup process for surface mounted components and pin components, from microscope setup, to what that gunk is, to temperatures etc... would be great to see and try to learn it!
Great work! have a look at the USB chip at 13:03..... near the R5? mark and R=330E (?pin 21, 22) The original soldering work is not connecting the chip to the pads. Not a propper soldering Maybe just reflowing the chip would have done the trick.
I think that was me poking around off screen, the probe just squashed the solder. But you are right, I could have tried re-flowing it first! Thankfully those IC's are cheap! Thanks for watching!
I repaired mine by applying a little heat to the USB chip. It has been running for 4 years. What I don't understand is why you changed the USB connector without first checking your connections. It seems that in this model, Scarlet 2i4, this failure is quite common. I subscribe to the channel because of your excellent videos
Hello and welcome! The USB connector was broken, the central part was missing and had the terminals just hanging in the air! Sorry if that wasn't clear in the video. Yes, I could have tried re-flowing that IC for sure. Not sure why I didn't try? Maybe to avoid stressing the PCB twice with heat. Some PCBs are not great and traces start detaching when heat is applied, even at low-ish temps. But I have to be honest and I don't think I had this elaborate thought back then! :D
You made everything complicated without reason, heat chip up and press it down. There was too much solder on the ground pad in the middle and that kept chip in air above PCB. Just press it next time, extra solder will pop up somewhere on side, just clean it with soldering iron and that's it.
27:49 solution for such miserable soldering is to take a magnifying glass and look sideways to ensure proper connection. Just take PCB in your palm and try to look like Sherlock Holmes ;) .Been there seen that, i was also frustrated about my hard soldering job ending up in flaky connections :) Don't be afraid to fry some chips to gain experience, i have a ton of broken stuff from equivalent of Craigslist, just to train my eye-hand coordination. Also camera microscope doesn't help as it gives you only 2d image where depth is being of essence ;) I personally have trained to recognize tip's blurryness as missing depth parameter so my wicked brain gets some sense of 3d while looking though my microscope setup. Last but not least - don't be afraid to push on QFNs while solder is melted. This ensures proper planar contact of flat chip to flat PCB; excess solder will come out as a ball to happily wick it off.
Thank you for all the hints! I did push the IC but maybe the ground pad was not molten yet. As you say, next one will be better! :) Thanks for watching!
I was impressed by your excellent failure analysis and repair. I like watching various people's repair videos, and I feel like I often see USB chip failures among them. On the contrary, there are almost no cases where the CPU/GPU core fails. Perhaps the amount of protection circuitry is different. thank you.
What Can cause a sound card to fail ?, sencondly I Can know if a part from sound card driver there is some thing else to add because I use Scarlett 18i8 1st génération in the computer from mixcontrol I see the output signal when I play music but that does not don't come out to the baffle
That not firmware - thats the eeprom that holds on the setup values for the firmware.The firmware is burned into the MCU's OTP-memory and that is read-out protected in regular case.What surprises me a bit is your working technique. You can desolder it like this. Cleaning the pads is also ok. Re-tinning is only necessary for the central PAD in the middle. For this I use solder past. There is still enough solder applied to the pads. A little flux applied to the pads is enough for now. Position the chip on the pads and then heat it with the hot air gun until the solder paste has melted. Solder the pin pads individually using a fine soldering tip, a little dripped flux and solder paste or very thin solder (0.5mm diameter). This works very well with the JBC soldering station. The JBC knife is too wide. The amount of flux they used here is simply too much. You won't be able to remove it from under the structures and in the long run it will attack the components and the circuit board. How conductivity changes over time is difficult to predict. Flux is good but it needs to be used as little as possible. The “no clean” products also leave residue. In the end it worked. The USB chip tends to die sometimes because it has no overvoltage protection at the input and if I saw the circuit correctly, there were no suppressor diodes at the input. Only the mantle wave filter was visible. Nevertheless, congratulations on a successful repair. I'm always as happy as a child at Christmas when I've managed a repair. As components become ever smaller, repairs are becoming more and more of a challenge.
I have the same model device and I suspect it has the exact same problem. It's been sitting in a box for at least a year, I never tried fixing it, and somehow the algorithm knew to recommend this video. I'm not so confident with my SMD soldering but I'll have to try fixing it now
I pulled two of these out of a bin behind a music store, both had broken USB sockets like that. I replaced them and all worked fine. I assumed at the time that the music store had vandalised them to prevent them from being re-used.
Since you are obviously a genius at this - there is a common problem with the Scarlett 2i2 where it boots on, flashes orange on both gain nobs a few times, is recognized by the OS but has NO sound at all and NO input signal with or without phantom power. Thoughts? Focusrite tech support say - broken, so sorry for your bad luck ;P
ahah no geniuses here! :) No, I've never heard of that. I suppose it might be a Firmware issue. If the interface doesn't have a way to re-load the firmware when the interface is not in a working status, it might require some PCB work. Does the I/F show ok on the OS? That might indicate the CPU inside the Scarlett is actually working so the issue is somewhere else. Sorry, not really familiar with that fault.
I don't mean to be nitpicking but that doesn't look like a card at all. 😜 I mean I think we usually call internal, PCI-E or PCI sound interfaces as soundcard, thn these are called usb audio interfaces. Sorry. Easing the pain of a pedant...
ahah no worries, you're not the first! I think I've been calling it alternatively card and interface throughout the video! Won't be the last time my English delivers funny things! Thanks for watching!
It's not too bad to replace and apparently it's a standard USB port. I have a feeling that the USB chip doesn't like those loose connections, the pins on my 2i2 were not bent or mangled.
1st time on this channel: - looks like you know what you do - I don't know if anyone british can confirm this, but you sound almost like a mothertongue speaker Anyhow: how did you learn all of these? I am entering my 40s and I really wanna learn this but can't find some good resources to do so.. Any suggestion?
Thank you! :) How did I learn? Electronics has always fascinated me since I was a kid. But I never received a proper academic education for a number of reasons. So I learnt by "tinkering" and - most importantly - by making mistakes and breaking things! :) I'm sure there is material available to give you some basics and nowadays tools, components and breadboards are very affordable and you can get your own "starting kit" very easily. Thanks for watching!
Well done ! Most DIY'ers would have deemed fake IC as the problem. There is a huge disbelief in IC's from China despite the fact that nearly all commercial available IC's are fabricated in China.
Well, there is also a huge amount of fake ICs coming from China as well! But yes, sometimes I think the same: what would be the monetary advantage to remove a similar IC, re-label it and sell it for pennies? I guess I should consider the almost-free labour and big sale numbers? Thanks for watching!
How do you find schematics! I bought a pre-Sonus interface and it's in great condition except for the power supply and the company is absolutely no help. At first I thought it was a burned out resistor but after replacing (had to guess known defect even the replacement was discolored so decoding impossible) that it's still not fully powering up. I bought another unit and swap the power supply over and everything is functional. So now no it's just the power supply I suspect maybe a capacitor because it's trying to start and then failing and then repeating.
I google! Sometimes I am lucky. Sometimes I am not. In this case I was lucky enough to find the 2i4 ones which are linked in the description if you are interested. Well done for fixing the Sonus!
23:34 I didn't realize they'd have pads not connected to anything. Guess it makes sense, just to hold the chip in place, but I was worried when that pad was gone lol
Yes it's not uncommon. Had the pad been connected, I guess I would have had to wear my telescope glasses and attempt a bodge wire :) So glad that that was not necessary! Thanks for watching!
I have a Focusrite 2i2 sound card, it was damaged due to a lightning strike, it has been repaired for maybe 2 times to my local services distributor and it was damaged again, is there a solution for that?
If your equipment tends to be damaged by lightings, a surge protected power strip - or better a UPS - might help. A friend used to have the same issue and resolved with an UPS. That said, the 2i2 is powered by USB. I'd expect that a surge would damage your motherboard before the USB connected interface? Unless the surge came from some attached sound devices? I don't know if there is a more general solution you could apply to your whole house. Thanks for watching!
I'm not sure showing this schematic 4:12 is allowed. Yes you can just google it but it's probably some leaked document, not an officially published document by the company. You can just tell the audience that you found a schematic without explicitly showing it in the video here on youtube. I hope you wont get sued or anything. Nice video by the way!
As you say it’s a document publicly available online and it’s been online for a few years :) If Focusrite are not happy they can contact me 🙂 Thanks for watching!
Awesome, from what I could see, it doesn't look like there is any ESD protection on the USB signals, big no-no and probably what went wrong with the original IC
Interesting! I find it ok for my needs - though I noticed some weird behaviour on the noise floor but intermittent so maybe it was just a glitch. What is it that you don't like of this interface? Thanks for watching and the 486 says hi :)
I had a similar device and it broke, and I sadly had to throw it away. These devices aren't very high quality, but for most beginning musicians and or producers they're very capable.
I'd agree. My MOTU is better but also 5 times the price :) Unfortunately those issues are seldom worth a repair unless you can do that by yourself. Thanks for watching!
Hello, nice repair for sure... One question, what are You spraying around 29:40 to all connectors, pots and switches? Some "magic" contact cleaner or just simple isopropylic alcohol to clean it a little bit??
Thank you! That is my favourite contact cleaner, it's called Servisol S10. It's "lubricated" and enhances conductivity. So it's not "dry", it leaves stuff behind. But I found that when the dry contact cleaner doesn't work, the S10 does :) I believe it's now sold by a different manufacturer - Kontakt 61 maybe? Thanks for watching!
The "Grime" you mentioned on the knobs is rubber coating disintegrating. I have an old Maudio interface which has been in a shelf for years and all the knobs are a sticky mess.
That was my first thought as well. Looked exactly like the rubberized muck I'm used to seeing on aging devices. Haven't seen it on knobs before though, I'll admit.
My first idea was that the 5V supply finds its way from the broken plug to the USB Data input and destroys it. I had several broken USB plugs that destroys the USB port in the past.
The 3343 USB chip is rated for a maximum of 5V on the data lines so that shouldn't damage it. Maybe some static discharge? Or maybe the IC didn't like the intermittent contact. Thanks for watching!
Heh, funny stuff. I did the exact same repair on the exact same card for the exact same reason a few years ago. I somehow managed to do this with almost no tools and little to no diagnostic, but there was a hint for me - the L4 was burnt up, so I just outright replaced the USB interface chip and shorted the inductor. Same issues with the pads and the same solution - just went around soldering them individually but I did remove that cap. I didn't even have a rework station at the time, had to use a small soldering torch instead of a heat gun. I guess those IC's like to go out when the port is damaged.
Well, a burnt inductor seems to suggest a big spike in there! Not surprised the USB IC was also dead! Well done, not sure I would have been able to do the same without proper tools!
Great video!! I am learning a lot :) - How do you plug the device while holding the oscilloscope probe? or, how many hands do you have? :) - I would love to see your setup
Thank you! One day I might make a video about my setup! Nothing special about the rest, the trick is to almost plug the USB plug, then set the oscilloscope, then use one hand to fully plug the USB connector. But yes, tricky. Especially when also filming on three cameras, looking at the scope, diagnose, think of what to say and recording a commentary :)
Ic chips intended for at least 2 heat cycles: first for manufactoring, and second for first time soldering. So, i y have enough expierence for not to burn pcb irself, it'll be normal
You should invest in some solder paste rather than using regular solder. That will make placing SMD ICs qa little easier, especially when they have a large ground plane like this chip.
Hi - I do have some, I didn't think of using it. The first time you do such a small chip, it's going to be a mess anyways :) Next time will be better! Thanks for watching!
Watched your video through my 1st Gen 2i2. To be fair - you do apply the goo just like Rossmann and he himself has mentioned that his hands are shaking and he rarely does the soldering himself anymore.
Hard to not to subscribe to you. A refreshing change instead of the guys that just seem to can’t get it wrong. At least you don’t hide your mistakes and errors behind the scenes and show the actual process of repairing stuff and how it goes on daily basis, not just some glorifying UA-cam video
hi... first hard solder is tuff.kkkkkk.. well... it need more solder in the pcb, you dont need to wick solder out, you just clean using new solder. and to help you can claen the new chip, and left a litle bit of solder, it will improve a lot. and finaly press down and the excess solder will go out on the border, that a sign thats all the chip is flat in pcb :) and you will never miss again. excellent work never the less. ps- i avoid to solder just one pad.... even if the rest apears ok, usse a blade and redone all the line at same time...
Thank you, I appreciate all the tips! I hope I can do another one soon! My blade tip was too big, I think you can see it at the beginning. It wouldn't even fit between the IC and the capacitor! I'll get a smaller one for next time! Thanks again!
Focusrite Scarlet units can be found in many broadcast radio and television environments including several BBC broadcast centres. So is capable of broadcast quality audio.
Once i replaced usb port on a focusrite and also computer did not recognize it.I think it has its own drivers through a dedicated software because many models have no stand alone operation and work only by its own software and no other way.Ableton hardware also had similar action after repair.
thanks and welcome! I've been using this for some audio tests and I'm pleased with it - besides one channel introducing some noise sometimes requiring a "reseat" but it could be the cheapo jack...
Good show! Very iNtErEsTiNg... I'm a new fan And thank you for speaking "to the" camera 🎥as apposed to "stage left" were most creators seem to be enamored with "their" own-image?.... It's a bit uncanny valley Cheers!
To me it looks like the IC might have been sanded and re-lasered, doesn't look like an original new part. Could be recovered from ewaste. As others said, that will make it harder to solder.
Could be. It came in tape though. For the price it feels not worth the time to prepare it but I know that labour there is not very expensive so who knows... I should have pre-soldered the pads as many others said! Thanks for watching!
I am very puzzled too. The XMOS datasheet says one of those pins are ground. Doesn't make sense but there is dust on those joints and with a microscope it's clear that those have never been touched. It's super weird but... it works :) Thanks for watching!
Really greatly enjoyable video Tony! many kudos to you! it's the first time I watch one of your videos after I've seen hundreds of others about electronic repair. What made me super surprised it's to realize about your italian origin and yet, till that point, I had never imagined that English isn't your native language. How did it happen that you speak English so fluently without the typical italian accent? ha? I envy you 😛non solo perchè parli inglese benissimo, ma anche perchè sei estremamente capace. Fare una riparazione elettronica è già una sfida che richiede capacità, ma farci un video come l'hai fatto tu, dettagliato, informativo, con grafica dedicata e composizione, ci vogliono ancora più competenze e lavoro... lo so perchè l'ho provato io stesso. Saluti da un Veneto scappato dal manicomio italia a 42 anni suonati e che da solo e senza mai esperienza di viaggi all'estero è andato dall'altra parte del mondo a cercare di lavorare e vivere in un paese dove la pazzia non ha ancora pervaso tutto e lo stato non è già diventato pervasivo, oppressivo e vessatore, e si può ancora vivere un po' come Dio vuole e non come vogliono una massa di psicopatici col coltello dalla parte del manico. Poi, se mi dici che flux usi e che stagno usi... mi fai un piacere perchè quello che ho comprato io fa schifo... ma ce ne sono mille in vendita e mica posso comprarli tutti per sapere quale è quello buono! 😉 Auguri Tony a di nuovo complimenti!
Hello and thanks for your comment! I've been working on my English for a while now, I always wanted to reach a level where my original nationality was not super-obvious! That said, I was told that my accent was good many years ago, so maybe I just like languages! I shall know soon, I'm trying to learn Portuguese! Clearly I am proud of my Florentine accent, ma magari ti sorprendera' sapere che sono Fiorentino solo per meta'. Per l'altra meta' sono perlappunto Veneto :) Ho automaticamente letto la parte Italiana del tuo commento con accento veneto! :) Capisco bene il motivo della tua decisione, e' un peccato, non e' vero? Grazie per le belle parole - come dici, fare questi video e' un lavorone ed e' sempre bello sapere che qualcuno li apprezza! Lo stagno uso il Loctite 309 60EN 5C - Non ho grande esperienza con altri ma questo secondo me funziona molto bene. Per il flussante, l'Amtech 559 e' il piu' gettonato ma e' molto caro. Quello che vedi nel video e' di Aliexpress, si chiama Kingbo RMA-218. In questo caso avrei dovuto usare l'Amtech probabilmente ma per roba generica il Kingbo da 3 sterline va benissimo secondo me! Grazie per la visita e per il commento, a presto!
What's the reason for using an USB-Soundbox like this? I see it as yet another annoying box with cables wasting space on your desk, or wherever. Isn't it way more efficient to simply use a digital output method like SPDIF or HDMI and leave the D/A-conversion do real dedicated HiFi-Hardware like the Amp instead?
I’d say microphone input first. Good quality output then. I’m testing sound devices and I need something quiet and accurate. The 2i2 is accurate enough, a laptop sound card would not. Also I need balanced inputs and outputs which the 2i2 offers. For general use it’s clearly unnecessary🙂 thanks for watching!
@@tony359 Thanks! Yes, I see. Input may be a reason. Otherwise you'd have to use an external A/D-converter instead, which probably wouldn't be advantageous.
The joy of a repairman rewarded by their work is infectious. As an aspiring electronics repairman who's stuck in a dead-end job, I hope to one day be half as good as you Tony. Magical work.
I know that feeling very well - I hope you can succeed in your dreams! Thanks for watching!
@@tony359 No, thank you Tony. Your channel has been an invaluable resource for amateurs like me. Also been entertaining. Your discipline is enviable. I would've thrown in the towel after the USB port lmao.
Indeed infectious. I was almost as happy as he was when he finally got the USB working.
ahahah thank you! :)
I really enjoyed this repair 👏👏
Thank you for watching! :)
I think it would have so much easier if you take that capacitor off first and put it back in after you're done soldering that chip
Yes, I regret not doing that indeed! Thanks for watching!
That stuff on the knob doesn't look like grime to me. More like what's left of TPU coating. It turns to goo and sticks to everything.
It could have been. I hope it was! But the whole device was literally covered in grime when I got it so... well, let's be optimistic and say it was TPU :) Thanks for watching!
@@tony359 I enjoyed it and learned something.
Woooh!
Great job with the repair, and even a greater job with the video editing, multimeter/oscilloscope shots, schematics and overall description of the whole process.
Always a pleasure and very informative.
Thank you so much, it means a lot to me! The editing takes soooooo long! :D
New QFN chips that have likely been sitting around for 10+ years tend to not stick unless you tin the pads under the chip first. So add flux and solder and tin the pads then when you hot air the chip it will automatically stick and all pads will be soldered. Also, you made the right call by wicking away most of the solder on the middle ground pad. Too much solder can squeeze out and create shorts on the outer pads which would be bad, most likely resulting in having to remove the chip and start over.
Thank you! Many have suggested the same! I shall do that next time! And good to hear that my decision to remove some of that solder was a good idea :)
apparently Louis Rossman is not the only one knowing how to solder tiny ICs!
@@alelondon23 Louis is actually not that good. Yeah to the below-average noob he looks like a god but vs a pro electronic engineer he's pretty average. I've publicly documented smd soldering repairs 10X harder than anything he's done. Look them up you might learn something lol
@@g4z-kb7ct I'd love to. link?
@@tony359 imho (and just for reference I'm a test engineer for a PCB assembly company so a lot of times I have to deal with soldering defects lol :P), you could have even removed all of the solder and still have gotten away with it. It's mostly used as a heatsink/ground pad and even the small amount of solder left there would have still made a good connection and besides it's a sound card, fairly low USB data transfer rate so the chip won't get hot. The most important thing with soldering QFNs is for them to lie absolutely flat on the PCB surface. Even minor level imbalances are annoying af to deal with. You should also have removed the capacitor from the side so that you could have better accessibility. It's a bulk decoupling capacitor, even if you ripped the pads you could still hack it back together and it would have worked :P
Your thought process troubleshooting was so great to learn from, thank you for keeping it in the video. And what a satisfying repair! I hope Focusrite see it and realise how important it is to share schematics and firmware versions for their community.
Well, from their point of view it’s one less 2i2 sold so I wouldn’t hold my breath 😂
Thanks for watching!
28:11 I had a moment like that when i replaced the HDMI connector on a PS4. Some of the pads were broken off. I spent hours with tiny strands of copper wire to repair the traces and it looked ugly. I was jumping up and down in excitement, shouting: "YES YES YES! It works! YES!" when it finally worked. I couldn't believe it. It was such a relief and I was so happy.
28:44 That smile on your face says it all.
I know the feeling! Those pads are a pain! I replaced one on the PS3 but it was still a through hole!
That 486dx2/66 casually laying on the bench. Anyway, great repair. Random Italian "cursing" when something unexpected happen is just icing on the cake. :)
ahah - but I did not curse! :D
That 486 was passing by, waiting for its motherboard to be fixed (coming soon when I get new tweezers!)
Thanks for watching!
Good eye. Took me ages to find it. Ah, I remember when they were the new frontier.
Great diagnostic work as per usual.
You might want to give solderpaste a go instead of reapplying solder to the pads.
That way you can skip soldering pin for pin.
Clean the pads, apply a small amount of solderpaste, put the chip in place and blast it with hot air.
I have some, I thought that solder was a bit easier for my first time. Who knows. I hope I can practice again soon! Thanks for watching!
@@tony359after some practice its alot easier.
Just get some of the cheapest chips you can find in the package you want to practice and some practice pcb's.
Recently got myself a bunch of 4 layer pcb's with big internal ground planes made just to practice soldering VQFN packages for an upcoming project.
Love your repair videos Tony, this one was a fun watch. Congrats on keeping the audio card out of land fill and hopefully saving some money too!
Thank you!
Ciao Tony. I was the service manager for Technics musical instruments for the entire life cycle of Technics. I love the way you explain what you are doing.
It's not easy for one to make a living repairing digital circuitry. Customers have no idea that we work at microscopic level.
I am an Italian national and your English is really brilliant. I heard the Italian long eeee sound where your pronunciation of the word ' this ' (singular) sounds exactly like ' these ' (plural). If you listen to an English speaker you will hear the word ' this ' sound a very short I sound with a long S thîssss and ' these' sounds like theeez.
Thanks for your comment, for watching and for your kind words!
My brain has a limited processing power, if I focus I think I can say "this" and "these" properly but if 'm doing three more things... well the language centre goes into default mode and then my Italian re-surfaces a bit! :)
At least I don't say "theess-e" :) (well, I hope I don't at least!)
yes. you can clearly hear the italian accent. but (at least to me) it's rather charming (at least much better than german accent in english (i am german)) 😁
@@BloodyClash danke schön!
Thank you to let us experience some real Italian temperamento😊
ahahah I contained myself! :D
Thanks for watching!
What a great, real-life video. I've always been a bit scared of SMT hand soldering and you've shown that I don't need to be. Thank you!
Through Hole is always simpler! But nice to have a challenge every now and then! Thanks for watching!
Muchas gracias por tu aporte.
En mi caso es una iTrack Solo que de la nada la pc dejó de reconocer la Interfaz
La solución que me funcionó fue usar la estación de calor en el circuito usb3343 para que se vuelvan a soldar los pads, no fue necesario cambiarlo
Dejo aquí el comentario por si a alguien le sirve :)
Thank you - someone else also mentioned that reflowing the USB3343 was all it was needed. Who knows, maybe it would have worked in my case too! Thanks for commenting!
You got a nice pick there... I always used MOTU soundcards, but for the money the Focusrite Scarlett was a formidable little soundcard especially for it's price point!
Indeed. I am using it regularly now. I still get my MOTU for more serious tasks but the 2i2 is pretty cool! Thanks for watching!
@@tony359 I watched a few videos once your channel came up in the algorithm. Very cool stuff, subscribed!
Amazing, thank you and welcome :)
Great work! Your frustration and joy were both palpable.
I can't imagine working on something so tiny with that USB chip. Well done!
Thank you! Yes, it was kind of scary to work on something so small! Thankfully I have a microscope!
The "YES" moment was worth it. A little luck together with know what´ya do can work out. Congrats and a happy new year wishes
Thank you, happy new year to you too!
Fantastic. Great work. I've worked on mice and keyboard for people in the past. The human slime factor is always gnarly with peripherals.
OMG, I can just imagine!
This was great! I'm secretly happy you didn't get this to work on the first attempt...proving you are human like the rest of us lol! I'm often working on vintage audio gear from the 1970s... really have to be careful with those solder pads as the bond between the board and the pad itself will fail even if you look at them funny. It was very rewarding to see this device come back to life. I really liked the segment where you pulled the firmware chip to validate. Fantastic as always!
thank you for your kind words! Yes, it was that good balance between "easy" and "challenging"! Just the socket would have been boring :) Thanks for watching!
you have to hold down the IC with your tweezers and add hot air when you put it on, otherwise great video! 😆
I did - 24:17. Clearly not enough heat? But I see the IC sliding a bit so I am a bit puzzled! But thanks for your input!
Great work dude. Amazed you didn't blow all those capacitors away accidentally when using the heat gun. I would have made a right mess 🤣
eheh, at least that went well! SMD is always tricky! Thank you!
28:33 It is a good practice to put some fresh solder on the IC itself not only on PCB pads, and I think you've forgot that. But overall, nice repair.
Very good idea! I actually removed some from the ground pad as I thought it would be too much? I have a feeling "more is better", the excess would squeeze out when you press the IC! Thanks for watching!
@@tony359 More isn't better, not even close, less is good, but the trouble is, this IC probably spent years nobody quite knows where, its entire bottom is likely oxidised and may not be eager to accept solder before you re-tin it. I think that was why you had unsoldered pins. As such this is indeed a good call. Maybe your flux is also not quite angry enough to deal with that without a little extra encouragement.
Serious manufacturers are highly reluctant to use ICs whose manufacture date is more than about 2 years old for this very reason, this is how they end up cheaply on the Huaqiangbei market.
very good point! I was indeed wondering how such an IC would work in a proper production line, in a reflow oven!
I've never done that with any IC. Drag-soldering with a chisel-point soldering tip and using plenty of flux works well.
Very difficult little chip of the devil there to solder indeed Tony... But glad you eventually persuaded it to stay at its seat completely, but the most important thing here is the correct troubleshooting method you used to discover that tiny faulty IC.
I was a bit scared with the firmware situation cause you did not say, do you run the card with 214 firmware or reverted back to the original that the little ROM had inside?
Also the black knobs seem to have that velvet feeling coating which eventually turns to a gross glue on plastics and looks dirty without being.
Cheers from Greece, keep up, best wishes and better year for all of us than that shitty 2023! Jim.
I did revert to the 2i2 firmware, it just didn't make it in the final cut! :)
Yes, that "grime" could have been that "soft rubber" which disintegrates over time. But I am not 100% sure, the whole thing was honestly covered in grime when I got it! :D
Thanks for watching!
Loved how you described in detail, what tool you use for what purpose in your hunt for the issue. Made it even more fun for me to follow along. Thank you!
Thank you for your kind words, very much appreciated!
I have the 2i4, and it's developed a slight static in the outputs when I change the volume knob. I've been too intimidated to try cleaning it, but watching you do this repair has shamed me into trying it. :)
Thos pots are more or less sealed. You can try with some cleaner but make sure you use the "lubricated" one, for trimpots. Failing that, you could replace it.
Good luck and thanks for watching!
That poor abused little red box! Well done for saving it!
Thank you!
Great work! Replacing this kind of devices has no monetary value (a working unit should go for around 30-40 USD second hand) but fixing any devices instead of buying a new device is always good for the environment.
Absolutely - thought a 2i2 here in the UK would sell for £60-70 ($82) so this was actually a good thing :) Thanks for watching!
Good investigation, analysis, and fix!
This reminds me of what my team and I were trying to do when we were setting up a streaming system during 2020/COVID. While we were pretty good at computers, we were still trying to figure out how to spell "audio-visual", and all the unfamiliar hardware bits and pieces to bring sound and picture together for a decent stream.
And just like you did here, it all eventually came together and it just worked!
Great job, Tony. I really enjoy your presentations - keep them coming
👍😁
Thank you for your kind words and for watching!
Nice work! Those J-lead style packages are such a hassle to solder, even for larger ICs, and because of the angle it's not so obvious when you have a bad joint.
Yes! But I can't wait to encounter another one to practice more! Thanks for watching!
Italian here. I can confirm a very polite, gentlemanlike, reaction to the desoldering incident 😅
ahaha grazie!
Just a minor correction: that isn’t a J-lead package. J-lead means packages that really have pins that come out the side and then fold down, with a curved end folded under the chip. The most common J-lead package is the PLCC (though not the version used for LEDs).
Thank you!
LOL you dont have to be rossman to know you need smd tip for soldering that ic package ! :D (8 years fixing electronics!)
I understand how you feel doing this type of work. I just did my first surface mount ic like this and if I were being paid they would have fired me after the second hour. 😂
ahah - well, someone has to begin from something! Thanks for watching!
Great job Tony, what is missing I personalny would like to see the communication between the new chip and CPU on an osciloscope.
Thank you, I appreciate the feedback! I'm always trying to find a balance between an enjoyable length and good content. Your feedback is important! Cheers!
I need to replace one of these on my Macbook pro and couldnt find a video for a quick tutorial to make sure was doing it best way possible, amazing video thank you! I never thought about soundcards using same USB IC controller.
oh yes, I wouldn't expect that either! Good luck with the repair and thank you for watching!
This USB port can be changed from ppl, that solder for one day. that was an easy fix.
Sorry for the lost pad - but it was toooo much heat for that little guy.
One day! If that was the case Alex at NorthRidgeFix would be without work in no time! :)
The headphone PCB on my 8i6 Mk1 needed similar work. Great product otherwise.
Yes, I'm pleased with it! Thanks for watching!
Nice when you spoke in Italian :-)
Solo nei momenti intensi 😂
The third gen focusrite I tested had a big peak in noise at 90KHz. A problem I noticed on a lot of other interfaces by other brands. While you can not hear it, it does effect the next opamp in the chain.
Interesting find. I suppose that could be easily filtered out with some passive components? Thanks for watching!
That's the shaped noise from the DAC. The CS4272 DAC is a relatively old and inexpensive design with like 1 or 2 bits only that requires fairly high-order noise shaping (4th or 5th-order I think) for it to achieve its rated performance, resulting in substantial amounts of ultrasonic noise that requires dedicated external filtering in addition to what's built into the chip... which in turn inexpensive interfaces will commonly skimp on, and at least the 2i4 is no exception.
The "big peak at 90 kHz" isn't even an actual peak, that's just the effect of ADC filtering kicking in when recording at 192 kHz. Things merrily keep rising beyond that point for a good while longer. Seems to be about a 3rd-order slope, so even the circuitry suggested by the datasheet would not be entirely adequate if you are serious about your hi-res.
@@PileOfEmptyTapes amazing insight, thank you!
@@PileOfEmptyTapes
The codec may be an old design, but it is also not understood. Otherwise someone would place an analog low pass filter on a higher end interface.
I ran a test where I mixed a guitar with a 32KHz sine into a distortion box. The poor opamp got so saturated reproducing the 32KHz tone that it undistorted the guitar.
It is my understanding that a codec works by sampling at a much higher rate and lower bit depth into an adder accumulator. (CIC filter) Then you carefully select a series flat-top and or half-band filters to push the noise out of band as much as you can while reducing the sampling rate. I guess you are saying that the noise is the result of inadequate digital filtering for an interface rated for 192 KHz and a lack of any analog filtering.
@@moddaudio "The poor opamp got so saturated reproducing the 32KHz tone that it undistorted the guitar." Congratulations, you have proven that high-frequency AC bias helps a baseband signal overcome a strong nonlinearity. Messrs. von Braunmühl and Weber really were onto something, eh? ;)
The importance of this effect for audio reproduction cannot be overstated. It helped tape overcome its scratchy lo-fi beginnings and turn into an actual hi-fi medium, and without it things like delta-sigma A/D and D/A converters (which are vastly dominating audio applications) or Class D audio amplifiers (which are using a switch-mode output stage and PWM) would not exist.
The CS4272/CS4392 presumably runs a dual bit modulator (3 levels or ~1.585 bits) at 128fs (12.288 MHz at 48 kHz in). Even including 21 dB of oversampling gain, it would only achieve a dynamic range of about 32 dB when applying flat (white-noise) dithering, absolutely pitiful performance. These kinds of modulators need substantial amounts of noise shaping to shift most of this noise outside of the audio bandwidth, which results in a noise floor rising with frequency - so you get less where it matters and more above. The more dynamic range you are trying to squeeze out of a given modulator and sample rate by increasing modulator order, the steeper the rise is. While there is some internal analog (switched-capacitor) filtering in the DAC to address ultrasonic noise output, it is unlikely to be comprehensive in a price-conscious DAC design in particular.
The maximum modulator order I've seen in a DAC has been 5th order, and when designers can get away with 4th or 3rd order while still achieving the desired performance they will happily do so, just because it makes them easier to manage. The maximum on the ADC side has been 7th order, which is really extreme but not such a big deal there as digital filtering can get rid of the noise during decimation. Even getting such a beast mathematically stable would have been quite the accomplishment (it used to be a real problem for the design of higher-order modulators).
At least running a higher-order modulator means that increasing the oversampling ratio (i.e. its operating frequency) gives major gains - at 4th order or a 24 dB/octave slope, doubling it turns a converter not even quite making it to CD quality spec into a midrange audio interface affair, assuming the analog side can keep up. A lot of designs made the jump from 64fs to 128fs in about 1995-98.
I realize this is a real rabbit hole into the nitty-gritty of digital audio. Understanding dithering with or without noise shaping is not trivial.
Fun fact, shaped dither was instrumental in the CD revolution itself. Remember that Philips only had a 14-bit DAC at their disposal (the TDA1540)? Using a digital filter IC to run it at 4X oversampling gave them another 6 dB, but that would still only have been 15 bits effective. So the filter would apply some noise-shaped dithering as well, which pushed performance levels in the audio band to about 16-bit level as desired. Mind you, the filter created some new problems - a 0.2 dB periodic passband ripple spec in an FIR filter is not pretty, for one.
Very nice. Do you have any videos on that whole flux/soldering/cleanup process for surface mounted components and pin components, from microscope setup, to what that gunk is, to temperatures etc... would be great to see and try to learn it!
Thank you! I've never considered making a video about that, thanks for the idea! It could be a useful one! Thanks for watching!
Great work! have a look at the USB chip at 13:03..... near the R5? mark and R=330E (?pin 21, 22)
The original soldering work is not connecting the chip to the pads. Not a propper soldering
Maybe just reflowing the chip would have done the trick.
I think that was me poking around off screen, the probe just squashed the solder. But you are right, I could have tried re-flowing it first! Thankfully those IC's are cheap! Thanks for watching!
I repaired mine by applying a little heat to the USB chip.
It has been running for 4 years.
What I don't understand is why you changed the USB connector without first checking your connections.
It seems that in this model, Scarlet 2i4, this failure is quite common.
I subscribe to the channel because of your excellent videos
Hello and welcome! The USB connector was broken, the central part was missing and had the terminals just hanging in the air! Sorry if that wasn't clear in the video.
Yes, I could have tried re-flowing that IC for sure. Not sure why I didn't try? Maybe to avoid stressing the PCB twice with heat. Some PCBs are not great and traces start detaching when heat is applied, even at low-ish temps. But I have to be honest and I don't think I had this elaborate thought back then! :D
You made everything complicated without reason, heat chip up and press it down. There was too much solder on the ground pad in the middle and that kept chip in air above PCB. Just press it next time, extra solder will pop up somewhere on side, just clean it with soldering iron and that's it.
I did press it, 24:18 ! :)
As I said, this was my first time with this kind of package, I'm learning! First time is not going to be perfect! :)
27:49 solution for such miserable soldering is to take a magnifying glass and look sideways to ensure proper connection. Just take PCB in your palm and try to look like Sherlock Holmes ;) .Been there seen that, i was also frustrated about my hard soldering job ending up in flaky connections :) Don't be afraid to fry some chips to gain experience, i have a ton of broken stuff from equivalent of Craigslist, just to train my eye-hand coordination.
Also camera microscope doesn't help as it gives you only 2d image where depth is being of essence ;) I personally have trained to recognize tip's blurryness as missing depth parameter so my wicked brain gets some sense of 3d while looking though my microscope setup.
Last but not least - don't be afraid to push on QFNs while solder is melted. This ensures proper planar contact of flat chip to flat PCB; excess solder will come out as a ball to happily wick it off.
Thank you for all the hints! I did push the IC but maybe the ground pad was not molten yet. As you say, next one will be better! :) Thanks for watching!
I was impressed by your excellent failure analysis and repair. I like watching various people's repair videos, and I feel like I often see USB chip failures among them. On the contrary, there are almost no cases where the CPU/GPU core fails. Perhaps the amount of protection circuitry is different. thank you.
You're welcome, thanks for watching!
What Can cause a sound card to fail ?, sencondly I Can know if a part from sound card driver there is some thing else to add because I use Scarlett 18i8 1st génération in the computer from mixcontrol I see the output signal when I play music but that does not don't come out to the baffle
I'm not sure I understand your question but it's hard to help at a distance I'm afraid. If your 18i8 has no sound output, I'm not sure I can help.
That not firmware - thats the eeprom that holds on the setup values for the firmware.The firmware is burned into the MCU's OTP-memory and that is read-out protected in regular case.What surprises me a bit is your working technique. You can desolder it like this. Cleaning the pads is also ok. Re-tinning is only necessary for the central PAD in the middle. For this I use solder past. There is still enough solder applied to the pads. A little flux applied to the pads is enough for now. Position the chip on the pads and then heat it with the hot air gun until the solder paste has melted. Solder the pin pads individually using a fine soldering tip, a little dripped flux and solder paste or very thin solder (0.5mm diameter). This works very well with the JBC soldering station. The JBC knife is too wide. The amount of flux they used here is simply too much. You won't be able to remove it from under the structures and in the long run it will attack the components and the circuit board. How conductivity changes over time is difficult to predict. Flux is good but it needs to be used as little as possible. The “no clean” products also leave residue.
In the end it worked. The USB chip tends to die sometimes because it has no overvoltage protection at the input and if I saw the circuit correctly, there were no suppressor diodes at the input. Only the mantle wave filter was visible.
Nevertheless, congratulations on a successful repair. I'm always as happy as a child at Christmas when I've managed a repair. As components become ever smaller, repairs are becoming more and more of a challenge.
Thanks for your comment and all the pointers!
I have the same model device and I suspect it has the exact same problem. It's been sitting in a box for at least a year, I never tried fixing it, and somehow the algorithm knew to recommend this video. I'm not so confident with my SMD soldering but I'll have to try fixing it now
Google knows :)
Good luck, I hope you can bring it back!
I pulled two of these out of a bin behind a music store, both had broken USB sockets like that. I replaced them and all worked fine.
I assumed at the time that the music store had vandalised them to prevent them from being re-used.
it could be, such a horrible practice! Well done!
Since you are obviously a genius at this - there is a common problem with the Scarlett 2i2 where it boots on, flashes orange on both gain nobs a few times, is recognized by the OS but has NO sound at all and NO input signal with or without phantom power. Thoughts? Focusrite tech support say - broken, so sorry for your bad luck ;P
ahah no geniuses here! :)
No, I've never heard of that. I suppose it might be a Firmware issue. If the interface doesn't have a way to re-load the firmware when the interface is not in a working status, it might require some PCB work.
Does the I/F show ok on the OS? That might indicate the CPU inside the Scarlett is actually working so the issue is somewhere else. Sorry, not really familiar with that fault.
I don't mean to be nitpicking but that doesn't look like a card at all. 😜 I mean I think we usually call internal, PCI-E or PCI sound interfaces as soundcard, thn these are called usb audio interfaces.
Sorry. Easing the pain of a pedant...
ahah no worries, you're not the first! I think I've been calling it alternatively card and interface throughout the video! Won't be the last time my English delivers funny things! Thanks for watching!
The USB port on my scarlett solo 1st gen broke in the exact same way, so far i've just been coping with random disconnections when i nudge my table.
It's not too bad to replace and apparently it's a standard USB port. I have a feeling that the USB chip doesn't like those loose connections, the pins on my 2i2 were not bent or mangled.
@@tony359 I mangled one of the pins a while ago and managed to bend it back with a sewing needle and tweezers, I should replace it at some point
1st time on this channel:
- looks like you know what you do
- I don't know if anyone british can confirm this, but you sound almost like a mothertongue speaker
Anyhow:
how did you learn all of these? I am entering my 40s and I really wanna learn this but can't find some good resources to do so.. Any suggestion?
Thank you! :)
How did I learn? Electronics has always fascinated me since I was a kid. But I never received a proper academic education for a number of reasons. So I learnt by "tinkering" and - most importantly - by making mistakes and breaking things! :)
I'm sure there is material available to give you some basics and nowadays tools, components and breadboards are very affordable and you can get your own "starting kit" very easily.
Thanks for watching!
Great job, Tony. Expert troubleshooting and repair. I've learned quite a bit from this video. Thanks so much for sharing!
You're very welcome - thanks for watching!
Well done !
Most DIY'ers would have deemed fake IC as the problem.
There is a huge disbelief in IC's from China despite the fact that nearly all commercial available IC's are fabricated in China.
Well, there is also a huge amount of fake ICs coming from China as well! But yes, sometimes I think the same: what would be the monetary advantage to remove a similar IC, re-label it and sell it for pennies? I guess I should consider the almost-free labour and big sale numbers? Thanks for watching!
How do you find schematics! I bought a pre-Sonus interface and it's in great condition except for the power supply and the company is absolutely no help. At first I thought it was a burned out resistor but after replacing (had to guess known defect even the replacement was discolored so decoding impossible) that it's still not fully powering up. I bought another unit and swap the power supply over and everything is functional. So now no it's just the power supply I suspect maybe a capacitor because it's trying to start and then failing and then repeating.
I google! Sometimes I am lucky. Sometimes I am not. In this case I was lucky enough to find the 2i4 ones which are linked in the description if you are interested.
Well done for fixing the Sonus!
23:34 I didn't realize they'd have pads not connected to anything. Guess it makes sense, just to hold the chip in place, but I was worried when that pad was gone lol
Yes it's not uncommon. Had the pad been connected, I guess I would have had to wear my telescope glasses and attempt a bodge wire :)
So glad that that was not necessary!
Thanks for watching!
you have made way! ITS THE SOLDERING!@
I have a Focusrite 2i2 sound card, it was damaged due to a lightning strike, it has been repaired for maybe 2 times to my local services distributor and it was damaged again, is there a solution for that?
If your equipment tends to be damaged by lightings, a surge protected power strip - or better a UPS - might help. A friend used to have the same issue and resolved with an UPS.
That said, the 2i2 is powered by USB. I'd expect that a surge would damage your motherboard before the USB connected interface? Unless the surge came from some attached sound devices? I don't know if there is a more general solution you could apply to your whole house. Thanks for watching!
I'm not sure showing this schematic 4:12 is allowed. Yes you can just google it but it's probably some leaked document, not an officially published document by the company.
You can just tell the audience that you found a schematic without explicitly showing it in the video here on youtube.
I hope you wont get sued or anything.
Nice video by the way!
As you say it’s a document publicly available online and it’s been online for a few years :) If Focusrite are not happy they can contact me 🙂
Thanks for watching!
Awesome, from what I could see, it doesn't look like there is any ESD protection on the USB signals, big no-no and probably what went wrong with the original IC
Very interesting, others have reported the same fault with this board. How would an ESD protection look? Thanks for watching!
yeah a first gen Focusrite Scarlett is the last choice for a sound card I would make.
Good job on the rescue though.
(loling at the 486DX2)
Interesting! I find it ok for my needs - though I noticed some weird behaviour on the noise floor but intermittent so maybe it was just a glitch. What is it that you don't like of this interface?
Thanks for watching and the 486 says hi :)
I had a similar device and it broke, and I sadly had to throw it away. These devices aren't very high quality, but for most beginning musicians and or producers they're very capable.
I'd agree. My MOTU is better but also 5 times the price :)
Unfortunately those issues are seldom worth a repair unless you can do that by yourself.
Thanks for watching!
Hello, nice repair for sure... One question, what are You spraying around 29:40 to all connectors, pots and switches? Some "magic" contact cleaner or just simple isopropylic alcohol to clean it a little bit??
Thank you! That is my favourite contact cleaner, it's called Servisol S10. It's "lubricated" and enhances conductivity. So it's not "dry", it leaves stuff behind. But I found that when the dry contact cleaner doesn't work, the S10 does :)
I believe it's now sold by a different manufacturer - Kontakt 61 maybe?
Thanks for watching!
The "Grime" you mentioned on the knobs is rubber coating disintegrating. I have an old Maudio interface which has been in a shelf for years and all the knobs are a sticky mess.
I thought about that - Let's hope it's that then! :) Thanks for watching!
That was my first thought as well. Looked exactly like the rubberized muck I'm used to seeing on aging devices. Haven't seen it on knobs before though, I'll admit.
My first idea was that the 5V supply finds its way from the broken plug to the USB Data input and destroys it. I had several broken USB plugs that destroys the USB port in the past.
The 3343 USB chip is rated for a maximum of 5V on the data lines so that shouldn't damage it. Maybe some static discharge? Or maybe the IC didn't like the intermittent contact. Thanks for watching!
You pretty much de-soldered one side of R105 while working on the QFN. Probably should go and touch that up.
uh, very good eye! I hadn't noticed that. It's barely soldered indeed! Thanks for spotting and for mentioning!
Heh, funny stuff. I did the exact same repair on the exact same card for the exact same reason a few years ago. I somehow managed to do this with almost no tools and little to no diagnostic, but there was a hint for me - the L4 was burnt up, so I just outright replaced the USB interface chip and shorted the inductor. Same issues with the pads and the same solution - just went around soldering them individually but I did remove that cap. I didn't even have a rework station at the time, had to use a small soldering torch instead of a heat gun. I guess those IC's like to go out when the port is damaged.
Well, a burnt inductor seems to suggest a big spike in there! Not surprised the USB IC was also dead! Well done, not sure I would have been able to do the same without proper tools!
just thinking??...did you leave the firmware on the flash for the 4 port version??...great video by the way.!
No, I went back to the 2i2 one - sorry I didn't show that part in the video! Thank you for watching!
Great video!! I am learning a lot :) - How do you plug the device while holding the oscilloscope probe? or, how many hands do you have? :) - I would love to see your setup
Thank you! One day I might make a video about my setup!
Nothing special about the rest, the trick is to almost plug the USB plug, then set the oscilloscope, then use one hand to fully plug the USB connector. But yes, tricky. Especially when also filming on three cameras, looking at the scope, diagnose, think of what to say and recording a commentary :)
Ic chips intended for at least 2 heat cycles: first for manufactoring, and second for first time soldering.
So, i y have enough expierence for not to burn pcb irself, it'll be normal
Absolutely. Thanks for your feedback!
You should invest in some solder paste rather than using regular solder. That will make placing SMD ICs qa little easier, especially when they have a large ground plane like this chip.
Hi - I do have some, I didn't think of using it.
The first time you do such a small chip, it's going to be a mess anyways :)
Next time will be better! Thanks for watching!
Watched your video through my 1st Gen 2i2. To be fair - you do apply the goo just like Rossmann and he himself has mentioned that his hands are shaking and he rarely does the soldering himself anymore.
I'm not an avid follower of his channel, I hope he's fine! Good to know I am at least applying the goo as he does :) Thanks for watching!
Great! No more Spiders! :) Ahahaha! This was a very nice one! Thank you! When do you Set up your Superthanks and/or Patreon?
The No More Spiders lives on my desk :)
Thanks for asking, I think when I reach 20K :)
Hard to not to subscribe to you. A refreshing change instead of the guys that just seem to can’t get it wrong. At least you don’t hide your mistakes and errors behind the scenes and show the actual process of repairing stuff and how it goes on daily basis, not just some glorifying UA-cam video
Thank you, I appreciate your kind words - and welcome! "The fun is in the journey", I also like when a video shows me the dead ends and the mistakes!
Man, that soldering looked challenging. Great that you got it working.
It's smaller than small! :D Thank you for watching!
hi... first hard solder is tuff.kkkkkk.. well... it need more solder in the pcb, you dont need to wick solder out, you just clean using new solder. and to help you can claen the new chip, and left a litle bit of solder, it will improve a lot. and finaly press down and the excess solder will go out on the border, that a sign thats all the chip is flat in pcb :) and you will never miss again. excellent work never the less. ps- i avoid to solder just one pad.... even if the rest apears ok, usse a blade and redone all the line at same time...
Thank you, I appreciate all the tips! I hope I can do another one soon! My blade tip was too big, I think you can see it at the beginning. It wouldn't even fit between the IC and the capacitor! I'll get a smaller one for next time!
Thanks again!
15:23: Shouting at the screen that your scope is in AC - LOL and you find it 1 second later - I guess my shouting worked ;-)
Oh, that was you! Thank you! ;)
That's not grime on the knobs, it's the vile soft-touch paint that degrades after a few years to a sticky mess!
Yes I thought the same - though the whole device was literally covered in grime so I'm like 50-50! :) Thanks for watching!
Focusrite Scarlet units can be found in many broadcast radio and television environments including several BBC broadcast centres. So is capable of broadcast quality audio.
Interesting! Given the price, I do like them :) Thanks for watching!
Mhm, interesting chef's cooking video😁 Some like it medium, others well done.😂 Nice Servicework 😉
I've baked things on this channel before :)
Once i replaced usb port on a focusrite and also computer did not recognize it.I think it has its own drivers through a dedicated software because many models have no stand alone operation and work only by its own software and no other way.Ableton hardware also had similar action after repair.
Interesting, some very professional ones do that. Thanks for watching!
During my apprenticeship, we always fanned out this SolderWick so that it was wider and we thought that it would hold more solder.
I guess it might work to an extend until solder is unable to "travel" from a strand to another maybe? Thanks for watching!
Man, IDK how you guys can deal with SMD without destroying the whole thing. I'm too clumsy for this.
Well, the destructive part happens off camera of course! :D
Thanks for watching!
You da man Tony. I am about to buy one of these 2i2 fourth gen. And now I know what they look like inside. Chers. I'm subed.
thanks and welcome! I've been using this for some audio tests and I'm pleased with it - besides one channel introducing some noise sometimes requiring a "reseat" but it could be the cheapo jack...
29:35 sorry but i couldn't help myself xD what kind of weirdo lost it's hamster's buttplug in sound card xD
ahahahah yes! :D
makes sense the USB interface chip died, the USB port broke and probably shorted 5V directly to the data signals
Yes even though those data lines are rated up to 5V - that's what the datasheet says - so I'm a bit puzzled. Thanks for watching!
Good show! Very iNtErEsTiNg... I'm a new fan And thank you for speaking "to the" camera 🎥as apposed to "stage left" were most creators seem to be enamored with "their" own-image?.... It's a bit uncanny valley Cheers!
ahah, I didn't realise that that was even a thing :) Thank you!
To me it looks like the IC might have been sanded and re-lasered, doesn't look like an original new part. Could be recovered from ewaste. As others said, that will make it harder to solder.
Could be. It came in tape though. For the price it feels not worth the time to prepare it but I know that labour there is not very expensive so who knows... I should have pre-soldered the pads as many others said! Thanks for watching!
Even though you got it working, I am suspicious about the 2 bridges on that XMOS chip.
I am very puzzled too. The XMOS datasheet says one of those pins are ground. Doesn't make sense but there is dust on those joints and with a microscope it's clear that those have never been touched. It's super weird but... it works :)
Thanks for watching!
Really greatly enjoyable video Tony! many kudos to you! it's the first time I watch one of your videos after I've seen hundreds of others about electronic repair. What made me super surprised it's to realize about your italian origin and yet, till that point, I had never imagined that English isn't your native language. How did it happen that you speak English so fluently without the typical italian accent? ha? I envy you 😛non solo perchè parli inglese benissimo, ma anche perchè sei estremamente capace. Fare una riparazione elettronica è già una sfida che richiede capacità, ma farci un video come l'hai fatto tu, dettagliato, informativo, con grafica dedicata e composizione, ci vogliono ancora più competenze e lavoro... lo so perchè l'ho provato io stesso.
Saluti da un Veneto scappato dal manicomio italia a 42 anni suonati e che da solo e senza mai esperienza di viaggi all'estero è andato dall'altra parte del mondo a cercare di lavorare e vivere in un paese dove la pazzia non ha ancora pervaso tutto e lo stato non è già diventato pervasivo, oppressivo e vessatore, e si può ancora vivere un po' come Dio vuole e non come vogliono una massa di psicopatici col coltello dalla parte del manico.
Poi, se mi dici che flux usi e che stagno usi... mi fai un piacere perchè quello che ho comprato io fa schifo... ma ce ne sono mille in vendita e mica posso comprarli tutti per sapere quale è quello buono! 😉
Auguri Tony a di nuovo complimenti!
Hello and thanks for your comment! I've been working on my English for a while now, I always wanted to reach a level where my original nationality was not super-obvious! That said, I was told that my accent was good many years ago, so maybe I just like languages! I shall know soon, I'm trying to learn Portuguese! Clearly I am proud of my Florentine accent, ma magari ti sorprendera' sapere che sono Fiorentino solo per meta'. Per l'altra meta' sono perlappunto Veneto :) Ho automaticamente letto la parte Italiana del tuo commento con accento veneto! :)
Capisco bene il motivo della tua decisione, e' un peccato, non e' vero?
Grazie per le belle parole - come dici, fare questi video e' un lavorone ed e' sempre bello sapere che qualcuno li apprezza!
Lo stagno uso il Loctite 309 60EN 5C - Non ho grande esperienza con altri ma questo secondo me funziona molto bene. Per il flussante, l'Amtech 559 e' il piu' gettonato ma e' molto caro. Quello che vedi nel video e' di Aliexpress, si chiama Kingbo RMA-218. In questo caso avrei dovuto usare l'Amtech probabilmente ma per roba generica il Kingbo da 3 sterline va benissimo secondo me!
Grazie per la visita e per il commento, a presto!
nice troubleshooting methodology. it appears all you need is a micro-soldering station. knowledge and tools are the key.
Thanks - meanwhile I purchased a 210 handle for my JBC. Now I need some more tips :)
Well done! Keep on repqiring! Got the newer model of this soundcard, awesome piece of kit for the price.
Yes, this old one is performing ok too! Thanks for watching!
What's the reason for using an USB-Soundbox like this? I see it as yet another annoying box with cables wasting space on your desk, or wherever. Isn't it way more efficient to simply use a digital output method like SPDIF or HDMI and leave the D/A-conversion do real dedicated HiFi-Hardware like the Amp instead?
I’d say microphone input first. Good quality output then. I’m testing sound devices and I need something quiet and accurate. The 2i2 is accurate enough, a laptop sound card would not. Also I need balanced inputs and outputs which the 2i2 offers. For general use it’s clearly unnecessary🙂 thanks for watching!
@@tony359 Thanks! Yes, I see. Input may be a reason. Otherwise you'd have to use an external A/D-converter instead, which probably wouldn't be advantageous.
Definitely need to use thinner gauge wick, but most importantly, do not dwell on the pads or they will lift!
Absolutely - It's just that I seldom work with tiny SMD so all my tools are the wrong size :) Thanks for watching!
You could have dropped the jack connector piece in small metallic pot like the surgeons do in movies when removing a bullet
Ahahaha that’s a great idea! Next time!
Great work mate! I just picked up a new addiction hahaa. Might need to start learning how to do this stuff haha.
Yes it can be an addiction, a very rewarding one! :)
Thanks for watching!
Those capacitrs are nothing compared to the ones on GPU's! KIA!!! Is all I have to say about that one!
One of the reasons I tend to repair older items! :) I can actually see the components! :)
28:18 What is Subspace relay 39 in device manager? That sounds interesting.
It's a subspace link with the Enterprise of course. (it's my headphones!)
Thanks for watching!
I ever fear to the QFN, I need fix one device having that but the solder was bad too XD, I need get solder paste to do it right.
It is an interesting package indeed! Thanks for watching!