Maker Pro
Maker Pro

Power supply problem

Hi,

I know my way around the inside of a computer, but am completely unfamiliar with anything using capacitors and resistors etc so am looking for some advice.

By way of explanation before I get to my question, I have an old Cambridge Azur 640H music server - basically a mini PC with supposedly high end hifi bits inside. It stores music on a hard drive. I've had it for quite a few years and have undertaken various repairs/replacements and have grown quite attached to it, and very familiar with it's internals. The unit basically comprises

Large circular mains transformer

CD unit (as used in most PC’s)

A digital-audio converter circuit board

A mini-ITX motherboard

A hard drive

Another 640H one came up for sale recently on Ebay as "spares or repairs" and fancying a challenge I bought it. Swapping bits between the good unit and the Ebay one, I've narrowed the problem down to what I think is the DAC unit which, besides doing the digital/audio conversion bit, also provides a power supply to the motherboard.(see photos below, one shows the 640H unit with the DAC unit removed, the second is a close up of the DAC unit) ).


IMG_4007rev.jpg

IMG_4013rev2.jpg



I have done some testing (photo) and I can measure the voltage from the DAC unit to the motherboard and for the good unit it is slightly over 12 volts and for the Ebay one it is slightly less. HOWEVER For the good unit I can get the motherboard to power up (CPU fan would spin). Also the CD drive would open/close. For the faulty DAC the measured voltage varied from just under 12 volts to just over. However, the motherboard would not power up (CPU fan would give half a spin at most) and the CD drive would not operate.


IMG_4010rev.jpg


The connection from the mains transformer to the DAC unit is actually 19 volts AC So it appears that the DAC unit is converting from 19 volts AC to 12 volts DC.

The second photo above shows the rear part of the DAC unit where the power comes in from the transformer. The black finned item is a large heat sink, and attached to it are 5 electrical items, each with 3 pins. Googling the part number on the item they appear to be some form of voltage regulator. My best guess is that one or more of these has failed. It's about the best guess I can come up with and makes a bit of sense, but any more educated and informed opinions would be appreciated!


Regards

Alan C
 
If it's a standard motherboard (looks like it) then where is the 5V supply it will require?

The 12V supply seems correct - supply voltages have allowable tolerances of up to 5% so anywhere from 11.4 to 12.6V would be fine - but without the 5V supply the mobo won't work at all.
 
If it's a standard motherboard (looks like it) then where is the 5V supply it will require?

The 12V supply seems correct - supply voltages have allowable tolerances of up to 5% so anywhere from 11.4 to 12.6V would be fine - but without the 5V supply the mobo won't work at all.

Hi, the 12 volt supply from the DAC unit feeds another small PSU. It plugs into a connector on the motherboard and provides the 5v supply via the connector. In the photo below, the black/white wires go to the DAC unit and theblack/red/yellow supply the hard drive and cd unit.

IMG_4015.JPG

I'd meant to say in my original post but didnt want to over-elaborate, my impression (and may sound stupid) is that while the suspect unit may have an indicated 12volts at the connector, as soon as a demand is placed it is not man enough to meet the demand. Sorry a very poor way of expressing it. But it's the only explanation I can come up with for why it won't power the motherboard.
 
As you say, the power supply does not seem able to supply sufficient current.

Give us the type numbers of the five devices attached to the heat sink. The current passes through these.
 
If the adapter board takes 12V and then feeds 12/5V into the mobo then there's some form of power conversion circuitry on the adapter board..... show us close up pics of both sides of the board please.

It is not possible to determine if the 12V supply is being dragged down or the 5V is doing this TO the 12V supply.

Does the 12V supply TO the adapter board stay stable if the mobo 12/5V cable is unplugged? If so then it's something on the adapter board that's dodgy.
 
Thanks for the replies,

Duke 37

Very difficult to read the type numbers because they are obscured. Not all identical, at least two are
CQ 526
MBR 2045CTG

Kelly's eye

Sorry, not sure what you mean by 12volt supply to the adapter board. The supply to the DAC board below is 18v AC from the mains transformer. From memory the 12v supply to the mobo from the DAC board doesnt change when unplugged, but would need to check to confirm tomorrow (had to put all kit away as other half not to keen on me using kitchen worktop as electronics bench !!)




IMG_4017.JPG



IMG_4018.JPG





Regards
 
Looks like a 'bulging' capacitor here - the one with 'crap' on it....... can you confirm?

If the top IS bulging - change it. What does the connector adjacent to it supply power to?

caps.png
 
Looks like a 'bulging' capacitor here - the one with 'crap' on it....... can you confirm?

If the top IS bulging - change it. What does the connector adjacent to it supply power to?

View attachment 35454

Yes, capacitor does look slightly dodgy and is marginally more convex than the one on the "good" DAC. Not sure what the connector does, I'd have to check. BUT don't think it's relevant to the problem as th only connectors which have been connected up during my test of both the "good" DAC" and the "faulty" one are the input from the mains transformer and th output to the motherboard PSU, so logically other connections are irrelevant ?
 
I've posted a photo above
Nope - THIS board...

board.png

You say this board gets 12V and plugs into the mobo to provide the 5V supply too..... the board MUST, therefore, have some conversion circuitry on it and I believe the mobo isn't booting because it either isn't getting the 5V it requires or the 12V supply is pulled low when THIS board is powered up.
 
Agreed, BUT as already said, with the good DAC supplying this PSU, the motherboard powers up fine, with the faulty DAC supplying this PSU the motherboard does not power up, so this PSU is not relevant to my problem.

Just occurred to me - might be one of those " why didn't you say that before" things. When I connect the cable from the DAC unit to the motherboard PSU, I check whether the motherboard will power up is by shorting 2 pins on the motherboard ( you probably know - there's a collecfion of pins used for power on, HD on, reset etc! Shorting the right pair of pins powers up the motherboard)
 
Temporary confusion resolved, normal service is resumed....

If the 12V, as measured, remains at 12V (or thereabouts) then it isn't a problem with the supply being 'man enough'. If the supply wasn't man enough the voltage would fall.

What can affect operation is excessive ripple (and it will be high frequency ripple since the 12V supply is derived by a SMPS (switched mode) device). You can't properly measure such ripple using standard meters - it requires an oscilloscope - but another effect of such ripple is the destruction of capacitors that end up 'bulged'......... oh, wait....:D

Check the bulged capacitor designations - does it say 85C or 105C on the side? They don't look like (to me) what I would call 'quality' capacitors at first glance and cost-cutting in this area is a common cause of equipment failure.

The SMPS device is obviously working as you wouldn't get 12V from it otherwise so the remainder of the board, as far as power is concerned, seems to be ok.

What markings are on the 7 devices screwed to the board?
 
Last edited:
Temporary confusion resolved, normal service is resumed....

If the 12V, as measured, remains at 12V (or thereabouts) then it isn't a problem with the supply being 'man enough'. If the supply wasn't man enough the voltage would fall.

What can affect operation is excessive ripple (and it will be high frequency ripple since the 12V supply is derived by a SMPS (switched mode) device). You can't properly measure such ripple using standard meters - it requires an oscilloscope - but another effect of such ripple is the destruction of capacitors that end up 'bulged'......... oh, wait....:D

Check the bulged capacitor designations - does it say 85C or 105C on the side? They don't look like (to me) what I would call 'quality' capacitors at first glance and cost-cutting in this area is a common cause of equipment failure.

The SMPS device is obviously working as you wouldn't get 12V from it otherwise so the remainder of the board, as far as power is concerned, seems to be ok.

What markings are on the 7 devices screwed to the board?


Re the capacitors, they are "Xunda" - which googling don't seem to have a good reputation and I've seen on other forums that Cambridge Audio have been accused of using cheap Chinese components. The biggest (brown) capacitor has a 105C rating, the medium size ones including the one with the slight discolouration are 85C.

As for the 7 devices, think you can just make out the markings on the photo below. From the descriptions above each I think they are part of the digital- audio part of the board?

IMG_4020red.jpg


With regard to voltage. I set up my test rig this morning, but this time I was only getting 6 volts from the DAC unit to the motherboard !! Something is obviously seriously wrong!

Finally - one question - as I said the power to the DAC board comes from the mains transfomer and I've measured this at 19 volts AC. Presumably something on the DAC board must be firstly converting this AC voltage to DC and then something else (the 5 devices attached to the heat sink?) then regulate it down to 12 volts. So - where does the AC to DC conversion happen.

THanks for your patience.
 
Yeah, 85C capacitors on boards that derive supplies by switched-mode means is poor practise and you'd benefit from changing them.

Interestingly, and pleasingly, the pcb actually shows what those 7 regulators actually supply power too! Left-to-right you have:

7805, 7805, LM317, 7805, 7805, 7815(?) and 7915. These deliver the following voltages:

+5V.....+5V....+??V....+5V....+5V....+15V...........-15V the (??) may be anything like 9V, 12V etc and is set by external components whereas the other regulators are 'fixed voltage' types. You can, if you wish, look up the pinouts and measure the input and output of each regulator (the 78** and 79** devices are in-gnd-out respectively) to prove they are all individually working - just don't short out the pins!!!

The AC-DC conversion happens on that board - the big 'coil' is the inductor used in the Switched-Mode method of conversion. At least one of the devices on the heatsink is a diode - this will rectify the 19V AC, the big capacitor will smooth it and one of the other devices on the heatsink will do the 'magic' to convert it to 12V.

If you're not getting a proper 12V then I'd go for a capacitor change. Most times a SMPS supply stops working it does so either spectacularly (blowing a fuse or itself to bits) or the more sophisticated devices have current limiting to prevent damage in the event of faults.

Try connecting a 12V bulb (rated at 12W or 25W - get one from a car! or a 50W bulb - you might have low voltage halogen lighting in the kitchen?) straight across the 12V feed from the DAC board. It 'should' still provide 12V under this level of load - if it won't then you'll have to let us know more about the actual device on the big heatsink and/or attend to potentially dud capacitors.
 
Try connecting a 12V bulb (rated at 12W or 25W - get one from a car! or a 50W bulb - you might have low voltage halogen lighting in the kitchen?) straight across the 12V feed from the DAC board. It 'should' still provide 12V under this level of load - if it won't then you'll have to let us know more about the actual device on the big heatsink and/or attend to potentially dud capacitors.

Tried a 4/21 tail/stop bulb - for the 4W connection got a dull continuous light, for the 21W connection it was brighter but almost flashing !!
 
The 4 watt side of the bulb only loads the supply to some 300mA (0.3A) whilst any problems may only occur at the higher current draw end. Even the 21W part of the bulb - guaranteed to be brighter so no indication to be assumed - only loads to just under 2A. That 'should' be enough to prove the capability of the 12V side as far as current limiting is concerned (try wacking TWO 21W bulbs on it) which leaves you facing the prospect of duff capacitors or a fault on the 'other end' of the supply lead.
 
The 4 watt side of the bulb only loads the supply to some 300mA (0.3A) whilst any problems may only occur at the higher current draw end. Even the 21W part of the bulb - guaranteed to be brighter so no indication to be assumed - only loads to just under 2A. That 'should' be enough to prove the capability of the 12V side as far as current limiting is concerned (try wacking TWO 21W bulbs on it) which leaves you facing the prospect of duff capacitors or a fault on the 'other end' of the supply lead.


What about the fact that when trying to power the 21w side of the light it was flashing/blinking - my electrically ignorant thought was that it was struggllng to supply enough current??

I've tried to do a quick check on the 5 devices attached to the heat sink, basically both sies of each device is 12 volts +/- 0.5 v, which seems rather odd. I can post the full results if you like?

I'd like to play about and change the capacitors - could you recommend an online source?

Thanks again
 
OK, I missed that bit about the light flashing (you said it was 'bright, almost flashing') - that's not right - unless the power supply is only intended to deliver 2A (certainly not enough to power the PC board).

Can you get the device numbers on ALL the five on the heatsink?
 
Top