Electronics > RF, Microwave, Ham Radio

Clean EMC 48 Volt power supply recommendation?

(1/2) > >>

Anyone know  of a EMI clean 48 Volt power supply? All the Meanwell products I have bought are horrendous QRM generators. Beads and filters wont stop the broadband transmitters that are built into these power supplies!

This is for a LDMOS amplifier. So something in the 50 amp class.

I have wasted so much much money that I am thinking of just using a un-regulated  Linear supply.

Wow!  That's a lot of power - >2000 watts.  I would suggest looking for surplus Telcom supplies.  I have a 5 amp 48 volt supply that was surplussed.  It was around $25 I think but it was in a 20 lb steel enclosure.  It cost about $30 to ship it.  Anyway, it's extremely clean on both ends.
I guess you'll need one that can be wired for 220V with that power requirement.

Good luck

Look for server supplies for Blade enclosures, they're often 50V and lots of amps. The HP ones are very well designed and clean, surprisingly so.

Richard Head:

Commercial switchmodes can be used but you will need to place them in another metal enclosure with additional  filtering (DM and CM) on the AC input and DC output. You will have to use components with a high self-resonant frequencies and very short leads. All control cables should be either eliminated (if possible) or filtered with an in-line inductor and feedthrough capacitors. Apply RF principles of screening and shielding. Ideally you should sweep the filter after building it to confirm a high insertion loss over a wide bandwidth with no funny resonances. For HF radios you only need to cover 1-30Mhz which is pretty easy. The commercial shielded bulkhead type IEC filters (Shaffner, Schurter etc) are generally reasonably good for the high frequency end. Keep the primary AC filter separated from the DC filter by locating them in separated compartments in a metal enclosure. Keep the AC and DC leads physically separated as much as possible to reduce coupling between them. Using these techniques I have reduced the common mode and differential mode noise to below the noise floor of the EMI measuring receiver, so it can be done.
You also mustn't upset the thermal design/airflow at all.  Most attempts I've seen to solve your problem are no more than a couple of ferrite rings hung on the mains and DC leads, with poor results.
Good luck.

Richard Head:
Telecom supplies on their own are not clean enough. They only have to comply with CISPR 22A (telecoms centers) or CISPR 22B if intended to be used in customer premises. Neither of these limits are low enough to not interfere with a nearby HF SSB radio. However, they can be cleaned up as I mentioned in my previous post. They all have active power factor correction which is a plus at the higher power levels (but worse EMI). Most have blind mating connectors at the rear which are used for both the AC and DC. These may need to be removed.


[0] Message Index

[#] Next page

There was an error while thanking
Go to full version