Commercial SMPS outputs

[Carlp]

I have a meanwell SMPS (S-145-24) that is capable of providing 6A at 24V. See: http://www.meanwell.com/search/s-145/S-145-spec.pdf. Using it in a SE mosfet design. The unit has 2 +24V screw terminals (also has 2 -24V terminals). I'm wondering if each terminal is able put out the max current (6A, obviously at the max only one terminal could do that), or if each terminal has a max current output of 3A. I see nothing about that in the datasheet, and no support documentation about it on their website. Anyone know about this?

Many thanks!

[DUG]

We use DIN mounted Weidmuller smps at work and some have the 4 terminals you mentioned. two + and two -.

To confirm:

Power the unit off, wait until the output is 0V

Measure the resistance between the two + outputs. Probably connected
Measure the resistance between the two - outputs. Probably connected

Just noticed the link to the data sheet (little slow today)

It is actually indicated in the data sheet on page two in the table: Terminal Pin No. Assignment

It will be 6A max for the output

[Carlp]

Thanks, DUG. I know which pins are +24 and which are -24. What I'm wondering is if I have a 5 amp load (+24V), do I have to split it between the two pins, or can I draw the full 5 amps from just one pin?

Cheers!

[ChrisA]

Originally Posted by Carlp

Thanks, DUG. I know which pins are +24 and which are -24. What I'm wondering is if I have a 5 amp load (+24V), do I have to split it between the two pins, or can I draw the full 5 amps from just one pin?

Cheers!

As was said above of the resistance between the two pins is very close to zero thenyou can take the full output from one pin. Or look on the back of the circit card and see of both pins are connected to the same trace. Likely they are.

But watch the connector pin itself. Some pins are rated at less than 6A.

[Carlp]

Well, I went ahead and tried connecting to one terminal and so far it's performed fine. In short, I converted my amp to a head amp, but was getting a ton of buzz, so someone suggested some serious CRC filtering. I didn't have parts on hand to filter 2 channels separately (each channel was powered by a separate terminal), hence my question. I'd need to pull one leg from the SMPS, filter it, then apply to each individual channel.

Since it's single-ended/Class A, I assume it's pretty much drawing the current it will draw at all times, so I think it has worked. No problems after a few minutes of testing, AND the big capacitance (about 17,000uF) has solved another problem - a turn-on thump.

Thanks guys.
Carl

[Carlp]

Well, not so fast.

For background, here's the amp I'm working on: DIY Class-A 2SK1058 MOSFET Amplifier Project. The CRC filter I added was 8200uF/100R/8200uF right at the output of the SMPS and before the 470uF/o.1uF caps on the signal schematic.

When listening longer, I realized I'm getting significant distortion at 1/3 volume and higher, and the volume level is low. I dug back in and found odd voltages. Where I had been getting 24V I was now reading about 9-10V - that's before the big load resistors and before the final 470/0.1uF caps. I wondered if it was because I was overdrawing the PS using only 1 terminal, but when I built 2 filters (a little smaller b/c I didn't have 4 of the BIG caps) I got the same result. So I tore off all the CRC filtering and I'm back in business - but with a noisy headphone output.

Any wisdom on why adding the filter would cause a substantial increase in distortion (basically, garbled, broken up sound)? I can't see how overdrawing the SMPS would do that, but can't figure out why else. It's as if the caps and 100R resistor I added were somehow burning off 15V or so.

Thanks for any ideas!

[ChrisA]

Originally Posted by Carlp

....
When listening longer, I realized I'm getting significant distortion at 1/3 volume and higher, and the volume level is low. I dug back in and found odd voltages. Where I had been getting 24V I was now reading about 9-10V - that's before the big load resistors and before the final 470/0.1uF caps. I wondered if it was because I was the PS using only 1 terminal, but when I built 2 filters (a little smaller b/c I didn't have 4 of the BIG caps) I got the same result. So I tore off all the CRC filtering and I'm back in business - but with a noisy headphone output.

Any wisdom on why adding the filter would cause a substantial increase in distortion (basically, garbled, broken up sound)? I can't see how overdrawing the SMPS would do that, but can't figure out why else. It's as if the caps and 100R resistor I added were somehow burning off 15V or so.

Thanks for any ideas!

Let's assume your guess is correct. Assume the 100R resister was
dropping 15 volts. Is this resonable? By Ohm's law we know V=IR.
Using your values we get
15 = I * 100 or
I = 15/100 or
i = 0.15 amps

So we see that if your amplifier were to pull 0.15 amps or 150
milliamps then you'd expect a 15 volt drop across the 100R resistor.
Your result seems reasonable.

A better design is to replace the 100R resistor with a choke. You
want the highest inductance choke you can afford that has very low DC
resistance (I'd say only a few ohms at most) and can handle the
required amount of current.

There are a lot of chokes available. One of these might work
http://www.hammondmfg.com/1540.htm

Or you can make a choke from a power transformers. Typically you can
use the primary as a choke and cut off the secondary lead wires.
First measure inductance and DCR and test current carrying ability

[Carlp]

Thanks Chris. I did the math, but obviously wrong as I got WAY over 0.15amps (not sure why, I wasn't too careful). So probably the supply voltage at the drain was far too low, and the load may not have seen much of a swing. That would explain the lower output levels. Would it explain the distortion? I don't know how to read mosfet data sheets (aside from the most basic of info).

[ChrisA]

Originally Posted by Carlp

... That would explain the lower output levels. Would it explain the distortion? I don't know how to read mosfet data sheets (aside from the most basic of info).

I'd expect the distortion to sound like soft clipping. Chokes work well because they don't drop many DC volts but can have very high reactance at the frequency you need to filter out.