Thanks for the feedback John! Much appreciated.
The reason they said to email Meanwell was just on the hunch that it would be made by them, but it wasn't certain.
I have excellent experiences with Meanwell power supplies, so I wouldn't even bother to save 13.50 USD on an 'unknown' unit. I'd say Teamacc is definitely right!
The reason they said to email Meanwell was just on the hunch that it would be made by them, but it wasn't certain.
I have excellent experiences with Meanwell power supplies, so I wouldn't even bother to save 13.50 USD on an 'unknown' unit. I'd say Teamacc is definitely right!
Meanwell comes set to 230v
I can’t believe I forgot to do this.
Well, I now know that when you buy stuff from China,
They don’t know what you’re going to do with it.
Remember that hum I have in my Class-D audio amplifier?
Well it ain’t there no more, thanks to a little switch on the
Power supply. This is the one that say 230/115V.
Slide it to 115V. Duh!
I can’t believe I forgot to do this.
Well, I now know that when you buy stuff from China,
They don’t know what you’re going to do with it.
Remember that hum I have in my Class-D audio amplifier?
Well it ain’t there no more, thanks to a little switch on the
Power supply. This is the one that say 230/115V.
Slide it to 115V. Duh!
Well, not that that speaks against it now...
Only I would myself buy something that handles mains voltage of a good well known brand, especially an SMPS since these are inherently less reliable. I see these supplies blow up frequently in our machine room at work where we have 67 servers and tons of other equipment running 24/7. We have airconditioning there large enough to create a BIG ice skating track and it barely keeps it cool enough. And we buy top of the bill equipment, most of them have solid state electrolytic capacitors (for a good reason!) If you understand the functioning of an SMPS, where mains is directly rectified, caps running at over 300Vdc voltages you'd understand it's risky business.
Choice is yours, I wouldn't feel good to have saved a few tenners and see my house burnt down one day...
Only I would myself buy something that handles mains voltage of a good well known brand, especially an SMPS since these are inherently less reliable. I see these supplies blow up frequently in our machine room at work where we have 67 servers and tons of other equipment running 24/7. We have airconditioning there large enough to create a BIG ice skating track and it barely keeps it cool enough. And we buy top of the bill equipment, most of them have solid state electrolytic capacitors (for a good reason!) If you understand the functioning of an SMPS, where mains is directly rectified, caps running at over 300Vdc voltages you'd understand it's risky business. Choice is yours, I wouldn't feel good to have saved a few tenners and see my house burnt down one day...
Connecting it elsewhere will not block the emission, this is not guaranteed to get rid of the noise. Your passive solution is rather sound though, not very elegant, but if it works it works...
But, what on earth did you place on the outputs? Are those gigantic zobels? Why?
Hello,
i've try to connect the fan ground the main ground,it is very efficient
but as you said,the best way is passive heatsink.
the new kits are comming with two heathsink,it's very easy to build it.
elsewhere,you can use pentium heatsink (with a saw to cut it,for insert it on the kit...
).thermal paste is correct ,comming with the stock stuff.no need to change it,i've chech the temp,it's OK.
obviously,a new coil with a low value go increase IDLE current.
i'm still listening with a linear regulated supply.it's ok.
yes,the Zobel are in a big size,(it's very easy to change values of capacitor or resistor ) ,the best listening is 470nF+10 nF.
i know that alower size have the same effects.
see you.
Attachments
Last edited:
I'm currently running a SureElectronics 4x100 amp with a 27V / 350W Meanwell SMPS and the Meanwell fan goes on every 15min or so for aprox 1min duration (and yes, it's VERY loud)... Someone suggested that I correct the DC offset on the board (which I've managed to get down to almost nil, by shorting the resistors) but it didn't make a difference, the fan still turns on regularly. Perhaps I've just got a faulty PS?
Hello,
i've try to connect the fan ground the main ground,it is very efficient
but as you said,the best way is passive heatsink.
the new kits are comming with two heathsink,it's very easy to build it.
elsewhere,you can use pentium heatsink (with a saw to cut it,for insert it on the kit...
yes,the Zobel are in a big size,it's very easy to change values of capacitor or resistor ) ,the best listening is 470nF+10 nF.
i know lower size have same effects.
see you.
I can't see why Sure hasn't connected it to the main ground by default?!
A passive heatsink is fine, but generally a designer evaluates the correct functioning of his design by measuring the temperature and when it is too high you connect a scope to the switching waveform to see what's going on. The temperature you guys are all mentioning makes me suspect there is something going on there, like I've said a few times now... It shouldn't get so terribly hot. I can cool my TK2050 amps with a thin and tiny slab of aluminium when I run them at 24Vdc! Can someone back this up? Kristleifur? You can place the whole world of aluminium on top to keep it cool, but if the cause of the problem is elsewhere why would you? In Dutch there's an expression for this, literally translated:"Mopping with the tap open..."
Then why on earth would you alter the zobel? Perhaps it gives a satisfying skewed frequency response, but the same applies here, when the problem lies elsewhere (usually in the speakers) why would you? You guys HAVE to understand the output filter has only little relationship to the audio quality itself. Sure it has an effect on the flatness of the response, but using audio grade components here is asking for other problems. Like emissions, one day you'll notice your TV starts distorting when the amplifier is turned on, or some wireless device is not working or going mad, or stuff just dies around you....
whatever.... This is HF, not audio. In HF a designer tries to keep things small and well accompanied by ground in a certain way to prevent EMC and EMI issues. Component choice is key to get the filter to even operate. This choice is rather NOT dominated by audiophile considerations. Like we don't want large PP caps there! Even the Sure designers understood well what to choose (and convenient for them the better choice here is often even cheaper!), hence I would even advise to keep this part unchanged. The ONLY improvement in the output filter, both in consumption, temperature and audio quality will be found in better inductors!But also here I say the choice is yours....
Even though I find some parts a bit exaggerated, yet he even could have added things on top which he stayed silent about in the article (compensates the exaggeration well
) here's a very good read:http://www.audioholics.com/educatio...ics/switching-amplifier-class-d-basics-page-2
Last edited:
Meanwell Fan
Yes, I think it may have a malfunction.
I use a single Meanwell driving two 4X100's and the fan has yet to come on.
My PS is is an oversized all metal enclosure. I am sure that helps.
And the two 4X100' boards are each in their own metal enclosure.
Time to check primary wiring and grounds!
Yes, I think it may have a malfunction.
I use a single Meanwell driving two 4X100's and the fan has yet to come on.
My PS is is an oversized all metal enclosure. I am sure that helps.
And the two 4X100' boards are each in their own metal enclosure.
Time to check primary wiring and grounds!
Last edited:
Hence my preaches to avoid anything with a fan and just build a plain and simple good 'ol linear supply!
But this is a bargain hunters thread, thus my preaching will probably be invain...
Really the idle current of this amplifier is probably enough to make the fan switch on. The offset trimmers can have a little effect, but not much. If it changes much then your speakers will have smoked out long before that. The trimmers on the sure board are single turn junk by the way, a good mod actually would be to mount some good Bourns multi turn trimmers there. At least then you can be certain that after playing loud a good rock track your offset will remain unchanged!
I have repaired a Passlabs X150 a short while ago which had single turn trimmers. I had the idle current and DC offsets trimmed nicely after breaking serious sweat and after I just gently tapped on the board I could start all over again. Pulled the suckers out and cuisined some nice 15 turn trimmers in, now it is running solid!
That said I am a bit afraid the Sure board doesn't allow for much larger trimmers, but Bourns have a miniature version as well....
Without proper examination of your Meanwell supply it's very hard to say why the fan is turning on and whether it is normal operation or not...A good start is to start at the load, measure the current consumption. Or an IR thermometer could perhaps provide some answers. Or maybe you can work around it by choosing a different fan. But then be sure the consumption is the same! And the air displacement! And of course the voltage!!
Last edited:
Edit: until resolved: Dont try
Found something:
According to the Tripath datasheet, the config pin (pin 24) should be connected to Vcc, whereas sure connects it to the ground.
With pin 24 to ground, the TK2050 acts as a stereo amp chip, whereas it is a bridged amp chip.
If this actually is something that might work, all credits go to elfishi, who first posted about him connecting pin 24 to Vcc
Edit: Until resolved: Dont try
Found something:
According to the Tripath datasheet, the config pin (pin 24) should be connected to Vcc, whereas sure connects it to the ground.
With pin 24 to ground, the TK2050 acts as a stereo amp chip, whereas it is a bridged amp chip.
If this actually is something that might work, all credits go to elfishi, who first posted about him connecting pin 24 to Vcc
Edit: Until resolved: Dont try
Last edited:
Found something:
According to the Tripath datasheet, the config pin (pin 24) should be connected to Vcc, whereas sure connects it to the ground.
With pin 24 to ground, the TK2050 acts as a stereo amp chip, whereas it is a bridged amp chip.
If this actually is something that might work, all credits go to elfishi, who first posted about him connecting pin 24 to Vcc
Edit: and if it doesnt work, blame elfishi
The board design is for non paralleled operation!! So NO don't connect Config to VDD I'd say. This would halve all series components in the output filter and place all parallel parts (like the zobel) offside!! It would turn the whole output filter in to a gigantic mess!!
EDIT: At hind sight the Sure board IS for parallelled TP2050 configuration, the Sure documentation present something that in my view wouldn't even work. If both bridges in each chip would work in counter phase (Config pin to GND, 'stereo' operation on the chip) then a board design for parallel operation would short circuit them out completely... Config would HAVE to be tied to VDD...
Last edited:
I can't see why Sure hasn't connected it to the main ground by default?!
it's so easy to use a comercial fan with its own wire.,this is a chinese amp,they won't add a couple of wire outside the board.this is a cheap amp,not the best design.
like the capacitor for analog low pass filter
A passive heatsink is fine, but generally a designer evaluates the correct functioning of his design by measuring the temperature and when it is too high you connect a scope to the switching waveform to see what's going on. The temperature you guys are all mentioning makes me suspect there is something going on there, like I've said a few times now... It shouldn't get so terribly hot. I can cool my TK2050 amps with a thin and tiny slab of aluminium when I run them at 24Vdc! Can someone back this up? Kristleifur? You can place the whole world of aluminium on top to keep it cool, but if the cause of the problem is elsewhere why would you? In Dutch there's an expression for this, literally translated:"Mopping with the tap open..."
ok,before build this amp,i've try a UCD topology (with post filter feedback)
i gave up this design because of bad THD,due to the layout,class D amp need SMD component.
the link
http://www.diyaudio.com/forums/class-d/159771-ir2110-failure-scope-probe.html
Then why on earth would you alter the zobel? Perhaps it gives a satisfying skewed frequency response, but the same applies here, when the problem lies elsewhere (usually in the speakers) why would you? You guys HAVE to understand the output filter has only little relationship to the audio quality itself. Sure it has an effect on the flatness of the response, but using audio grade components here is asking for other problems. Like emissions, one day you'll notice your TV starts distorting when the amplifier is turned on, or some wireless device is not working or going mad, or stuff just dies around you....
sorry,i wrote something false,zobel is 470nF+10 Ohms.
i own a digital scope meter,the SURE kit do not have a zobel on the output.
you can check it easily.
But also here I say the choice is yours....
Even though I find some parts a bit exaggerated, yet he even could have added things on top which he stayed silent about in the article (compensates the exaggeration well
as i said,in this case,component size are very easy for me to exchange,the size do not deteminate the quality of the amp.
i won't take the soldering station for 2 SMD components.
next i will post some THD and IMD measurement.
Last edited:
Any other input on the two different boards? I think the 4*100W looks good. Can I use only 2*100W on that board, will the other 2 channels still draw power if unconnected?
I'd like to avoid the fan version if possible.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.