I've had this thing in for a couple weeks, gotta wait for the firewire controller IC to arrive. For any of you who aren't familiar with this, it's a realtime firewire DSP, mixer and mic/instrument preamp. It costs ~$900 nowadays, when this one was bought 5 years ago it was $3k. And it was a factory refurb to begin with, so it isn't exactly a cheap device.
Failure cause is a common one for this model. The owner borrowed it to someone who plugged the firewire cable in backwards, frying the controller. I have some other suspicions about this (as plugging the firewire in backwards swaps the +30v supply with ground, something which shouldn't cause a failure, since the controller IC is 3.3v so obviously powered by a regulator and protected by diodes).
But what's shocking is that i found the main filter cap bulged and leaking, which is an Elna 35v 1000uF, no mention of low ESR anywhere. Somebody should tell them that Elna doesn't make SMPS caps. The other caps are from Nichicon and Meritek. They look okay but NONE of them are properly pushed into the board and they wiggle around. Nice job guys.
I was bored today so i also took apart the power supply because it was buzzing and whining. What i found inside horrified me. I bet this didn't cost more than $0.50 to make. Caps are all Jun Fu, with the primary side cap leaking. Since the cap inside the device itself is gone, i'll also test the secondary caps, they could very well have gone open. There is NO CONTROLLER IC, it's a self-oscillating flyback with optocoupler feedback. Oh yeah and THE MOV IS MISSING. Is this a power supply for a $3k device??? Jeez.
And don't let me get into what caused the failure in the first place - the bottom of the Firewire ports isn't soldered to the board, which allows the aluminum guard to split apart. Everybody i asked about this told me "no way it was plugged in reverse, it's foolproof!" but i was surprised about how easy it actually is to plug the cable the wrong way around in this thing. This is an unacceptable design fail. And all it would take to fix would've been a blob of solder.
Sorry for the pic quality btw, i forgot where i put my camera.
Edit: Primary cap tests OPEN. I threw it on the scope just for fun, and... 200mV p-p 50Hz ripple, 300mV p-p switching ripple and 600mV p-p high frequency ringing, most likely from the transistor switching spike. Those numbers are with the power supply UNLOADED. And consider that it actually goes thru a pi filter. I think this can truly be called "made by HongKongFlyApart, LLC".
Edit 2: The 600mV spikes were coming from my cellphone charger nearby. I unplugged that and now there is "just" 400mV. Doesn't make it less bad however.
Failure cause is a common one for this model. The owner borrowed it to someone who plugged the firewire cable in backwards, frying the controller. I have some other suspicions about this (as plugging the firewire in backwards swaps the +30v supply with ground, something which shouldn't cause a failure, since the controller IC is 3.3v so obviously powered by a regulator and protected by diodes).
But what's shocking is that i found the main filter cap bulged and leaking, which is an Elna 35v 1000uF, no mention of low ESR anywhere. Somebody should tell them that Elna doesn't make SMPS caps. The other caps are from Nichicon and Meritek. They look okay but NONE of them are properly pushed into the board and they wiggle around. Nice job guys.
I was bored today so i also took apart the power supply because it was buzzing and whining. What i found inside horrified me. I bet this didn't cost more than $0.50 to make. Caps are all Jun Fu, with the primary side cap leaking. Since the cap inside the device itself is gone, i'll also test the secondary caps, they could very well have gone open. There is NO CONTROLLER IC, it's a self-oscillating flyback with optocoupler feedback. Oh yeah and THE MOV IS MISSING. Is this a power supply for a $3k device??? Jeez.
And don't let me get into what caused the failure in the first place - the bottom of the Firewire ports isn't soldered to the board, which allows the aluminum guard to split apart. Everybody i asked about this told me "no way it was plugged in reverse, it's foolproof!" but i was surprised about how easy it actually is to plug the cable the wrong way around in this thing. This is an unacceptable design fail. And all it would take to fix would've been a blob of solder.
Sorry for the pic quality btw, i forgot where i put my camera.
Edit: Primary cap tests OPEN. I threw it on the scope just for fun, and... 200mV p-p 50Hz ripple, 300mV p-p switching ripple and 600mV p-p high frequency ringing, most likely from the transistor switching spike. Those numbers are with the power supply UNLOADED. And consider that it actually goes thru a pi filter. I think this can truly be called "made by HongKongFlyApart, LLC".
Edit 2: The 600mV spikes were coming from my cellphone charger nearby. I unplugged that and now there is "just" 400mV. Doesn't make it less bad however.
Comment