Really appreciate all your insights MorbidFractal, it's really helpfulI guess life is hard when you don't have the time to "roll your own", because you obviously can, and are left posting links to a random selection of E-Bay supplies and having to ask if they are any good.
Yes at least the FETs and maybe also the driver ... not easy to fix if so as it is molded into a plastic "box".You have probably blown the mosfets on the SMPS.
Possibly the driver too.
You could try getting a replacement and just say it broke on its own ?
I think I just have to purchase a new as it was obviously my own fault .... had reversed a power diode on the amp board!! from Vss/-V to an internal -5V ... shooting through to GND .... must have been tired when mounting the components
The other chanel built at the same time works like a charm
So rebuliding the burned amp right now.
... and this time testing on a much lower voltage with PTC fuses on the supply lines
Good to know. Apologies for being acerbic. BTW...
http://www.ti.com/lit/ds/slusc18a/slusc18a.pdf
The location of the sense resistor is perfectly valid given it is the upper switch that sets the input current to the bridge. The lower switch effectively just free wheels that current.
Of course if your supply has placed that resistor in the wrong place, source of the lower switch, then it is measuring free wheeling current and whilst still feasible would be sub-optimal.
http://www.ti.com/lit/ds/slusc18a/slusc18a.pdf
The location of the sense resistor is perfectly valid given it is the upper switch that sets the input current to the bridge. The lower switch effectively just free wheels that current.
Of course if your supply has placed that resistor in the wrong place, source of the lower switch, then it is measuring free wheeling current and whilst still feasible would be sub-optimal.
Attachments
Ok seen the information on Hypex (refered to by Reactance) and no real evidence it is a Hypex design .... so sorry, way too fast a conclusion on Hypex
Look at the silkscreen, then put some effort in by overlaying it with an SMPS1200 its 1:1 identical. Also the recommend SMPS for the Ncore would obviously be an SMPS800 or 1200 as the designer recommend, the logic protection board from the NCORE that drives the protection scheme for OV, OC and OT interfaces with THAT SMPS.
Last edited:
Latest update:
As written before, I had one more of the SMPS, this one just at +-70V.
I have now fixed amps, and tested well before I put it onto the 70V PSU
It runs fine.
Voltage is a little higher than written, but this is of course at a very low amp output, so close to idle .... here it is around +-76V
Only thing is that if I turn it of and then on again right after or not long after, the out put seems to do something strange, and it results in the amp making bad noises and at 70V it's quite a lot :| ...... I'm thinking that there is not enough bleeding in probably both output and on the primary side capacitors. (seems there is none on either side).
Will look further into this.
pronk have you experienced anything like this with yours?
As written before, I had one more of the SMPS, this one just at +-70V.
I have now fixed amps, and tested well before I put it onto the 70V PSU
It runs fine.
Voltage is a little higher than written, but this is of course at a very low amp output, so close to idle .... here it is around +-76V
Only thing is that if I turn it of and then on again right after or not long after, the out put seems to do something strange, and it results in the amp making bad noises and at 70V it's quite a lot :| ...... I'm thinking that there is not enough bleeding in probably both output and on the primary side capacitors. (seems there is none on either side).
Will look further into this.
pronk have you experienced anything like this with yours?
pronk have you experienced anything like this with yours?
There's definitely no bleeder on the output. I haven't noticed this problem with any of my builds. They are all class AB with an idle current of 90 to 130ma so the output voltage drops off rapidly after AC power is disconnected.
I also use speaker protection that disconnects the speakers if V+ is more than about 5V below the expected value so I likely would not hear any power supply induced start-up or shutdown noise.
Are your bad noises persistent or just a short burst? It certainly is possible that one rail comes up much more quickly than the other if output capacitor discharging was uneven. Just speculation but some amplifiers behave strangely if the rail voltages are too different. A pair of bleeder resistors would fix this problem.
It's like tunring on and of fast and repeatedly .... and it sounds like dc on the output .... just tured off immediately ..... so no chance to measure or anything ... only guessing
The amps have not shown anything like this on other psus .... notmally they just shut down and turn on very quitely .... no noise what so ever ....
Bleading resistors are jus a guess, and I would say a safety feature/requirement for the input caps
.... ok will investigate some more
The amps have not shown anything like this on other psus .... notmally they just shut down and turn on very quitely .... no noise what so ever ....
Bleading resistors are jus a guess, and I would say a safety feature/requirement for the input caps
.... ok will investigate some more
Hi
I did some degooping on the "black secret module" soldered, with some sharp blade and hot air it was pretty easy.
you can see pictures here : SMPS - Google Drive
unfortunately secrets remain secrets as soic8 IC were sanded...
I tried degooping because that module is the hottest part of the entire board, perhaps it will helps a bit.
I did some degooping on the "black secret module" soldered, with some sharp blade and hot air it was pretty easy.
you can see pictures here : SMPS - Google Drive
unfortunately secrets remain secrets as soic8 IC were sanded...
I tried degooping because that module is the hottest part of the entire board, perhaps it will helps a bit.
Hi t-minik
Interesting .... year so-8 .... could probably a few different drivers ... IR2153 could be a good guess
Did some more test, and amp now schwitches on and of no problem.
Problem was in the amp, where I have never tested it with as high voltage as her with +-75V in idle.
I have an under voltage detection which turns the driver off.
The amp needs to get the 12V driver voltage and the +-5V supply fully powered before driver is turned on. So in essense the undervoltage detection needs to trigger some 10 - 20 V below full Vcc = 75V in this case.
Changed Zenner to 66V, and it works like a charm.
Bassically I think the SMPS seems to work quite fine. So a bargin for what it delivers
It is not foolproof ... so don't make too many mistakes like me
Interesting .... year so-8 .... could probably a few different drivers ... IR2153 could be a good guess
Did some more test, and amp now schwitches on and of no problem.
Problem was in the amp, where I have never tested it with as high voltage as her with +-75V in idle.
I have an under voltage detection which turns the driver off.
The amp needs to get the 12V driver voltage and the +-5V supply fully powered before driver is turned on. So in essense the undervoltage detection needs to trigger some 10 - 20 V below full Vcc = 75V in this case.
Changed Zenner to 66V, and it works like a charm.
Bassically I think the SMPS seems to work quite fine. So a bargin for what it delivers
It is not foolproof ... so don't make too many mistakes like me
Last edited:
Had to change the output coil of the amp as it was running very hot with now +-75V, which leads to considerably more core loss than with previously up to +-50V .... so instead of one T106-2 core, I've glued two together stacking them with fewer windings .... now it is only getting margianlly warm in idle mode ...
Next is to build it all into an enclusure and test at max power
Next is to build it all into an enclusure and test at max power
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Power Supplies
- Amplifier SMPS on eBay ... any experience?