Well, the product documentation spends whole paragraphs lavishing in excruciating detail about ZVS, soft switching, low EMI, etc...
Like, buy me, buy me, buy me...
That's why I purchased this item and I'm a little bit disappointed about the huge noise spikes in the output you know... Possibly the target customer doesn't own a scope, or only a 10 MHz one so he cant see the spikes?...
I do not know enough about this technology to decide whether this is a small fixable bug in the design, or an epic screwup. Besides, they no give schematics.
I hope yu3ma can make his supply available as a kit, because the design looks like a BIG WINNER, I mean look at his output ripple measurements!
Right now I wouldn't attempt to build it, as that would require learning a lot about this tech. While that would no doubt be an extremely interesting endeavour, I don't have time... and also I need a supply for my amplifier like... about this week. Because my previous amp let out the magic smoke, so I'm without any decent amp at the moment.
So while I'm always for OVER-ENGINEERING THE HECK OUT OF STUFF, I guess what's gonna happen is that I'm gonna grab a toroid from the junk pile which is gonna be 10 volts too high, unwind the secondary a bit to adjust and call it a day.
Like, buy me, buy me, buy me...
That's why I purchased this item and I'm a little bit disappointed about the huge noise spikes in the output you know... Possibly the target customer doesn't own a scope, or only a 10 MHz one so he cant see the spikes?...
I do not know enough about this technology to decide whether this is a small fixable bug in the design, or an epic screwup. Besides, they no give schematics.
I hope yu3ma can make his supply available as a kit, because the design looks like a BIG WINNER, I mean look at his output ripple measurements!
Right now I wouldn't attempt to build it, as that would require learning a lot about this tech. While that would no doubt be an extremely interesting endeavour, I don't have time... and also I need a supply for my amplifier like... about this week. Because my previous amp let out the magic smoke, so I'm without any decent amp at the moment.
So while I'm always for OVER-ENGINEERING THE HECK OUT OF STUFF, I guess what's gonna happen is that I'm gonna grab a toroid from the junk pile which is gonna be 10 volts too high, unwind the secondary a bit to adjust and call it a day.
Sad to say, but that might be your way.
Btw technical standards of connex electronics smps never looked "professionnal" to me.
just my 2c
Allow me to chime in and reply to some of your questions.
Starting with metal enclosure.
This power supply is intended to be installed along with an amplifier in a metallic enclosure. Using a separate enclosure only for the power supply would defy its purpose of having extremely small footprint compared with any alternatives available on the market. The SMPS enclosure would have to be large enough to not create safety hazards because of close proximity links between primary and secondary which the enclosure would bring, affect the proper cooling unless the semiconductors would be thermally connected to it and would just make it look like any other $.99 incl. shipping Moneywell clone available on taliexpress or other similar places.
I still can't see the way how the probes were connected to SMPS to measure the ripple so I have reserves in taking the measurements results as absolute correct.
We all know (or not, since I get emails from time emails showing terrible ripple measurements results including worst possible probe setups) that ripple measurement has to be performed carefully using no GND leads, decoupling caps and no loops otherwise terrible result are obtained.
The copper shield is installed on the series available SMPS240QR's, and many other models except the resonant ones which for obvious reasons no need or can use it. Is still a mystery why this one didn't had one (lost during shipping
?) as we were shipped them with EMI shield. I should remind you that this is still not a LLC resonant supply, so all the comparison with an LLC supply and expected results are at least exaggerated if not wrong.
Thank you for your test and review, I eagerly expect to see your own SMPS built test results and compare them.
My
Thanks for taking the time to reply!
> I still can't see the way how the probes were connected to
> SMPS to measure the ripple so I have reserves in taking the
> measurements results as absolute correct.
No probes. As you say the ground wire with alligator clip is a sure way to fail. I simply soldered a coax directly at the output, with 50R source termination using a MELF SMD resistor. I like this method for fast signals, it works well. Next best thing is using the tiny ground spring on the probe, which gave similar results.
However the tiny ground spring tends to slip, and I don't like stuff slipping when there is a mains-powered board on the bench, so I soldered everything. It frees up a hand for pushing on the "take screenshot" button.
> The copper shield is installed on the series available SMPS240QR's
I got it from there:
SMPS240QR Module d'Alimentation Dcoupage 240W +/-30V - Audiophonics
No copper shield on the picture. Perhaps this is old stock? Anyway, if you ship it with copper shield now, then good!
> I should remind you that this is still not a LLC resonant supply, so all the
> comparison with an LLC supply and expected results are at least exaggerated
> if not wrong.
I think it's a marketing/documentation problem rather than a hardware/design problem. The supply works and stays cool. It does its job well for the price.
However, let's have a look at the docs:
http://www.connexelectronic.com/documents/SMPS240QR.pdf
(please note absence of copper shield on picture...)
"SMPS240QR use state of the art, highly efficient Series Resonant Converter Topology. Due to the soft-switched topology used, the SMPS240QR has very low EMI noise, lower losses and is much more compact than any similar power rating Switched Mode Power Supply, which most of them use classic hard-switched topology."
My gripe is that it promises "very low EMI noise" and well, this is not the case.
It promises high-efficiency and small form factor, and does deliver on that.
I believe there is a LLC resonant 300W version (SMPS300RE). It is larger and a bit more expensive, but still smaller and much lighter than a toroid of similar power. If the docs had clearly mentioned the noise difference between the two, I would most definitely have spent extra for the LLC, and perhaps I'd have been quite satisfied with it. Who knows?
> Thank you for your test and review, I eagerly expect to see
> your own SMPS built test results and compare them.
I'm certainly not going to invest the R&D time required to design and build one. I extracted a suitable toroid from the junk heap, it will do nicely!
> I still can't see the way how the probes were connected to
> SMPS to measure the ripple so I have reserves in taking the
> measurements results as absolute correct.
No probes. As you say the ground wire with alligator clip is a sure way to fail. I simply soldered a coax directly at the output, with 50R source termination using a MELF SMD resistor. I like this method for fast signals, it works well. Next best thing is using the tiny ground spring on the probe, which gave similar results.
However the tiny ground spring tends to slip, and I don't like stuff slipping when there is a mains-powered board on the bench, so I soldered everything. It frees up a hand for pushing on the "take screenshot" button.
> The copper shield is installed on the series available SMPS240QR's
I got it from there:
SMPS240QR Module d'Alimentation Dcoupage 240W +/-30V - Audiophonics
No copper shield on the picture. Perhaps this is old stock? Anyway, if you ship it with copper shield now, then good!
> I should remind you that this is still not a LLC resonant supply, so all the
> comparison with an LLC supply and expected results are at least exaggerated
> if not wrong.
I think it's a marketing/documentation problem rather than a hardware/design problem. The supply works and stays cool. It does its job well for the price.
However, let's have a look at the docs:
http://www.connexelectronic.com/documents/SMPS240QR.pdf
(please note absence of copper shield on picture...)
"SMPS240QR use state of the art, highly efficient Series Resonant Converter Topology. Due to the soft-switched topology used, the SMPS240QR has very low EMI noise, lower losses and is much more compact than any similar power rating Switched Mode Power Supply, which most of them use classic hard-switched topology."
My gripe is that it promises "very low EMI noise" and well, this is not the case.
It promises high-efficiency and small form factor, and does deliver on that.
I believe there is a LLC resonant 300W version (SMPS300RE). It is larger and a bit more expensive, but still smaller and much lighter than a toroid of similar power. If the docs had clearly mentioned the noise difference between the two, I would most definitely have spent extra for the LLC, and perhaps I'd have been quite satisfied with it. Who knows?
> Thank you for your test and review, I eagerly expect to see
> your own SMPS built test results and compare them.
I'm certainly not going to invest the R&D time required to design and build one. I extracted a suitable toroid from the junk heap, it will do nicely!
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.