Hi there!
I've had my trusty 0404 for ages. Used it for a long time on my PC but then got fed up with driver issues on WIN10 and bought a Focusrite instead. I was always really happy with the way the 0404 sounded though.. it was really good both on headphones and speakers.
Couple days ago I decided to bring the device back to use on another system but it died on me. At first I managed to get it powered on and working normally but it was really picky about the way the DC adaptor plug was seated inside the unit. I would get it to work only when I twisted the adaptor plug/cord to an angle. When I released the cord, power would cut out. I did what any lazy guy would, taped the plug in the right angle on top of the unit and enjoyed the ride. But now after a day, I can't get it powered on at all. The poor connection must've damaged something inside.
I opened the unit to see if the female plug on the unit had come loose from the board, but I'm reading 5 volts on the female plug pins. I resoldered the pins, but this didn't do anything. I checked the power switch for continuity and it seems to be ok.
How do I need to measure voltage to see if the unit is getting power? Most logical thing would be that the DC socket has failed but I don't understand why I'm reading normal 5v on the pins. I tried another DC power supply but this didn't help either. I would really appreciate it if anyone could point me in the right direction. Many thanks!
I've had my trusty 0404 for ages. Used it for a long time on my PC but then got fed up with driver issues on WIN10 and bought a Focusrite instead. I was always really happy with the way the 0404 sounded though.. it was really good both on headphones and speakers.
Couple days ago I decided to bring the device back to use on another system but it died on me. At first I managed to get it powered on and working normally but it was really picky about the way the DC adaptor plug was seated inside the unit. I would get it to work only when I twisted the adaptor plug/cord to an angle. When I released the cord, power would cut out. I did what any lazy guy would, taped the plug in the right angle on top of the unit and enjoyed the ride. But now after a day, I can't get it powered on at all. The poor connection must've damaged something inside.
I opened the unit to see if the female plug on the unit had come loose from the board, but I'm reading 5 volts on the female plug pins. I resoldered the pins, but this didn't do anything. I checked the power switch for continuity and it seems to be ok.
How do I need to measure voltage to see if the unit is getting power? Most logical thing would be that the DC socket has failed but I don't understand why I'm reading normal 5v on the pins. I tried another DC power supply but this didn't help either. I would really appreciate it if anyone could point me in the right direction. Many thanks!
I'm guessing no lights at all come on? The power switch is on the main board - does it get power, given the 5V connector is on the top pcb ?
There are schematics on-line in a few threads for the input circuitry, and various photos to identify parts related to input circuitry. You could identify some Vcc, Vee, gnd, and +5V and +48V (switched) points and see if they are powered. You could try and identify where the switched power input goes on the board - such as a dc/dc section for opamp Vcc/Vee.
There are schematics on-line in a few threads for the input circuitry, and various photos to identify parts related to input circuitry. You could identify some Vcc, Vee, gnd, and +5V and +48V (switched) points and see if they are powered. You could try and identify where the switched power input goes on the board - such as a dc/dc section for opamp Vcc/Vee.
Thanks a lot for the reply and sorry for taking time to answer back! No lights come on the card at all. Power switch gets 5v normally (I checked it with two boards together). I've tried to understand some of the schematics but haven't had much luck so far. I really don't know how to find the right spots to measure voltage. Plus the split board is a pain, I haven't got the cable connectors to have the boards powered while separated.
Yeh, you've got a tough job.
By inspection with the boards apart, if you can find a part or pad that is a known Vcc, or Vee, or +5V, then you may be able to carefully solder a flying lead to it, and then re-install the two halves and apply power and measure the voltage on the flying lead or leads. Given you have the known schematics of the front end,then perhaps try and find the larger parts like the large coupling e-caps - they should then get you to the associated opamps and Vcc and Vee, and gnd and +5V (using the schematic 'sdiy-left_input.gif'). The +48V is from a boost converter with an identifiable boost inductor, and there is likely a 5V feed going to that circuitry.
Also the noise reduction mod has a photo that identifies the 'dc-dc U23', which also likely has a 5V feed. Those 5V feeds may connect back to the switch, or to a device that connects to the switch.
You may be able to buzz out where the switched power rail goes, as far as the next parts to identify that could be faulty or missing a control signal to turn them on. Or the front panel LEDs get their power from a part that may be powered from the +5V in.
My notes on the +48V are as follows - perhaps of no use, but here they are:
48V phantom power distributes to each channel via large smt R44 and R59 - it is filtered by large C42 (100uF 63V). Connects to 100R R96 to 0R link R104 and via. R104 to bjt Q19 (likely phantom power switch) to nearby boost switchmode Q15/C44/L3 where Q15 is marked 4850 (Vishay SI4850 fet) - appears to be controlled via U10. Disabling that dc/dc (so that it didn't switch or load the 5V feed) could be as easy as removing the feed 5V but that appears to be an internal layer, so perhaps removing U10 and L5 parts is simplest. Sidiy's mod schematic shows the large SMT 48V feed resistors R44, R59, as well as the large 47uF 63V coupling caps C26, C27 were removed.
By inspection with the boards apart, if you can find a part or pad that is a known Vcc, or Vee, or +5V, then you may be able to carefully solder a flying lead to it, and then re-install the two halves and apply power and measure the voltage on the flying lead or leads. Given you have the known schematics of the front end,then perhaps try and find the larger parts like the large coupling e-caps - they should then get you to the associated opamps and Vcc and Vee, and gnd and +5V (using the schematic 'sdiy-left_input.gif'). The +48V is from a boost converter with an identifiable boost inductor, and there is likely a 5V feed going to that circuitry.
Also the noise reduction mod has a photo that identifies the 'dc-dc U23', which also likely has a 5V feed. Those 5V feeds may connect back to the switch, or to a device that connects to the switch.
You may be able to buzz out where the switched power rail goes, as far as the next parts to identify that could be faulty or missing a control signal to turn them on. Or the front panel LEDs get their power from a part that may be powered from the +5V in.
My notes on the +48V are as follows - perhaps of no use, but here they are:
48V phantom power distributes to each channel via large smt R44 and R59 - it is filtered by large C42 (100uF 63V). Connects to 100R R96 to 0R link R104 and via. R104 to bjt Q19 (likely phantom power switch) to nearby boost switchmode Q15/C44/L3 where Q15 is marked 4850 (Vishay SI4850 fet) - appears to be controlled via U10. Disabling that dc/dc (so that it didn't switch or load the 5V feed) could be as easy as removing the feed 5V but that appears to be an internal layer, so perhaps removing U10 and L5 parts is simplest. Sidiy's mod schematic shows the large SMT 48V feed resistors R44, R59, as well as the large 47uF 63V coupling caps C26, C27 were removed.
Thanks again for your help! And I tried to understand the last +48V part of it, but my brain went into error mode. 😀
I took a closer look at the board and might have made some progress since I found a component labeled U42 that has discoloration on the solder pads and maybe some fry marks on the component as well. On this thread someone else also had this component fry on their unit:
https://www.diyaudio.com/community/threads/e-mu-0404-usb-mod-wiki.162512/post-4685597
The part is also mentioned on the first post here.
https://www.diyaudio.com/community/threads/e-mu-0404-usb-mod-wiki.162512/
Could you possibly help me out how to test if the component is fried? Or could it be just a cold solder?
I took a closer look at the board and might have made some progress since I found a component labeled U42 that has discoloration on the solder pads and maybe some fry marks on the component as well. On this thread someone else also had this component fry on their unit:
https://www.diyaudio.com/community/threads/e-mu-0404-usb-mod-wiki.162512/post-4685597
The part is also mentioned on the first post here.
https://www.diyaudio.com/community/threads/e-mu-0404-usb-mod-wiki.162512/
Could you possibly help me out how to test if the component is fried? Or could it be just a cold solder?
It looks like that 5-pin part is TPS62200DBV, as the PHKI marking is consistent with datasheet. If you can probe +5V at pin 1 (across pins 1 and 2/gnd, and likely across C209), but get no 3.3V output (perhaps on other side of choke to gnd - which should also be across one of the smd caps). Use the datasheet from TI to identify pins and likely parts.
The concern would be that the 5V incoming is somehow switched upstream, and that switch is not working or not enabled, and so you can't energise U42 for starters to do any testing.
Well done for keeping going. It can be difficult to probe parts, so pointy probes can help check continuity between pads.
TP4 and the free pad area near U42 (to the right of the 3 terminal side of U42) may be good spots to solder wires to extend out for metering, once you confirm what circuit node they are.
If you can confirm that part is not working, then I guess you're half way to jumping the first hurdle.
The concern would be that the 5V incoming is somehow switched upstream, and that switch is not working or not enabled, and so you can't energise U42 for starters to do any testing.
Well done for keeping going. It can be difficult to probe parts, so pointy probes can help check continuity between pads.
TP4 and the free pad area near U42 (to the right of the 3 terminal side of U42) may be good spots to solder wires to extend out for metering, once you confirm what circuit node they are.
If you can confirm that part is not working, then I guess you're half way to jumping the first hurdle.
I just measured U42 as per your instructions. Pin 1 -> gnd gives +5V but pin 4 -> gnd gives 0V. Does this conclude that the part is fried?
The 'choke' is the inductor part marked with '100' and 'M04', and has two terminals at its corners that can be probed to check that the internal winding still measures a low resistance, and hasn't open-circuited due to an over-current - this would be a very unlikely fault, but is at least easy to measure.
Finally able to update this thread. My 0404 is alive again! A new PHKI chip from Aliexpress took ages to arrive but today I swapped U42 with a new one. Soldering was a bit of a pain (first time SMD soldering) but I managed it. I'm testing the card right now and it seems everything is working just perfect.
Thanks so much to @trobbins and @aliarifat794 for your help, I had no chance to complete repair without it!
If someone stumbles into this thread later, I had to get 10 pcs of the PHKI chip so I have spares. Drop me a message!
Thanks so much to @trobbins and @aliarifat794 for your help, I had no chance to complete repair without it!
If someone stumbles into this thread later, I had to get 10 pcs of the PHKI chip so I have spares. Drop me a message!