SMPS recommendations

[JohnsonAV]

Hey everyone,

This is my first post here, so go easy on me. I'm putting together two Class D Audio SDS-258's, what would be an appropriate power supply for such a combination? I spoke with Tom and he recommended +-65V, 8-10A. Any recommendations? I'm not looking to build one.

Thanks.

Aaron

[ssanmor]

You can have a look at our line of SMPS:
www.coldamp.com
I would recommend SPS80HV, with that product you can get above 1KW continuous. If you don't find what you need we can create custom voltages for you.

[fb]

Connexelectronic

Make sure to request the correct AC input voltage.

[ChocoHolic]

Of course also hypex might be a worth a look.
SMPS1200
The beauty of the smps1200 is the synchronous rectification, which improves efficacy , and even more valuable prevents all issues that are often discussed regarding supply rail pumping of class D half bridges.
IMHO also nice.
- An auxiliary supply: +/-12V
- A floating auxiliary supply for the lower gate drive and the bootstrap supply.

I do have one on my desk and different from multiple other brands the hypex smps passed my private visual safety check regarding creepages and clearances. Well - for some reason I did not unwind the transformer, but I did not see obvious short comings in the visible areas of the smps construction. So I got trust on the transformer as well.
Don't take me as a certification institute.
It's just that I do have a personal opinion (EN60065 minded) what I want to use in my living room, because sueing after death is not a fortunate option.

[JohnsonAV]

Thanks for the recommendations. I am trying to keep the height of the PSU under 44mm. So far it looks like the PCBstuff A1000SMPS is the only one in that form factor.

Any thoughts on this unit?

[ChocoHolic]

Johnson, you are bringing me into trouble!
....I am going to make friends now...

Con:
Look at the pic of the a1000smps.
When you look at the design you will find the power Mos (primary) and the output diodes (secondary) on the same heat sink. The way of isolation of the diodes lets expect almost no creepage vs the heat sink, while the way of isolation of the MosFets lets expect approx. 1.5mm creepage. So on the heat sink primary vs secondary have approx 1.5mm creepage. In Europe safety standards would need something between 6mm-8mm, depending on some details and interpretation of the EN60065. Another place of short creepage is on the left hand side of the black caps (primary) and the resistor (secondary) next to it.

Cold:
Again look to the pictures.
Well, they have special isolation for the MosFets, but distances between PCB vs. heat sink and between leads vs. heat sink appear pretty small.
Also the fact that you can see the copper of the windings in the transformer instead of margin tape does not generate trust to me.

For my living room both are out of the race.
Cold and Con: Feel free to tell me bad ...

P.S.
In fact in the meanwhile I am convinced that lots of the DIY SMPS stuff has short comings in safety. So there is no clue in avoiding Cold or Con while buying something else which might be worse!
There is almost no control from authorities regarding these safety topics.
It is up to us DIYers to check carefully what we have bought.
If unable to check, stay with products which have been approved by a valid certification institute and shop at a place where you can easily approach the vendor.

[JohnsonAV]

Thanks for the advice ChocoHolic, that's good to know. If only Hypex made a SMPS module that is less than 44mm in height.

[ssanmor]

Mmm, ChocoHolic, you only "make friends" if you don't expose the facts fairly and with real data.
Let me correct you in that "Cold" PSUs (you can say the complete name, please, we'll continue being friends) DO meet all the EN60065 safety regulations.
In fact our PSUs have more creepage and clearance than really required, and we even use TIW (triple insulated wire) for the transformer primary windings, a kind that is very expensive and usually reserved to medical specified units. In any case our units are Class-I devices requiring Earth connection.
It is not casual then that some big companies are using our supplies with total confidence and passing certifications with no issues.

So you can be confident about the safety of our PSU units, of course.

[ChocoHolic]

Hi Ssanmor,
glad to see you answering in a friendly and technical manner.
Protection class I and Tex-E (or similar triple insulated) wires...
I agree these wires would basically allow a transformer design with less or even without margin tape/spacers. Definitely not normal to use such wires for a big transformer in consumer applications. 'You rich???'
Class I devices - that's a key information.
Means your SMPS does not need to fullfil double/reinforced insulation, but basic insulation only. ...bringing down the required creepages by factor two vs. what I had anticipated. This could fit to the pictures.
Less demanding for the smps, but more demanding for the DIYer to fullfill the stringend requirements for all protective earth wirings.
Also more demanding for a proper system set up which does not suffer from earth loops - especially in distributed systems... (a desktop computer at one power outlet and the amp at another, the PS3 again somewhere else and a beamer... and may be even a radio with antenna plug ? ... pretty inconvenient to do audio in the class I world.., of course I always found a proper solution, but .. )

I have to apologize. I was automatically anticipating suitability for protection class II and furtheron would never have expected triple insulated wires . Nevertheless, your smps is still remaining out of the race for my private living room projects, because I decided to do my DIY audio stuff in protection class II.


I guess Cristi will show up soon here, too.
Looking forward to get his feedback about their safety concept.

[Cristi]

Originally Posted by ChocoHolic

Con:
Look at the pic of the a1000smps.
When you look at the design you will find the power Mos (primary) and the output diodes (secondary) on the same heat sink. The way of isolation of the diodes lets expect almost no creepage vs the heat sink, while the way of isolation of the MosFets lets expect approx. 1.5mm creepage. So on the heat sink primary vs secondary have approx 1.5mm creepage. In Europe safety standards would need something between 6mm-8mm, depending on some details and interpretation of the EN60065. Another place of short creepage is on the left hand side of the black caps (primary) and the resistor (secondary) next to it.

Perhaps you don't have the "full picture" when you make this statements.
Creapage between the primary switches and heatsink is 6.4 mm, the footprint is offset about 3.5mm inwards and the legs are bended to allow placing insulating material all the way down to the pcb. moreover, the heatsink or heatslug is connected to protective earth, so if any short should ocure, this will be between the transistors and protective earth connection. I'm not the one who invented the common heatsink for power devices, primary switches and secondary diodes, it is extensively used in industry since the first designs of SMPS's.
Next, the distance between the cap and the resistor is about 5mm. the capacitor sleeve can withstand at 1000V RMS insulation test. and to create an arch, one would need a lightning strike, in which case the power supply isolation is the last thing to worry about. I have tested the unit for safety, and, surprisingly has passed, and i'm about to make a full certification, as soon as will be required by an OEM or wholesale customer, right now 12-15k usd, certification costs are planned for something else for the moment. and btw, i wonder how many commercial smps for audio have the certifications done, not just according....
I believe that the next step will be to criticize the SMPS800R because this has the low profile which the topic starter was looking for. Connexelectronic if isn't obvious from the picture, the smallest clearance and creapage between primary and secondary side is on the transformer, and measures 6.6mm. the transistors and diodes are completely isolated from the heatsink.