What can I use a 13.8 V 20 A linear supply for?

[Siwastaja]

While it's certainly possible to have a switch-mode supply totally clean enough to power ham radios, I totally understand the desire to use something which is old, well-known and simple and kind of proven not to cause the kind of interference that many (especially poorly designed) switchers do cause. And efficiency is not so important to many. The energy spent in radios is mostly dissipated as heat anyway (some of course goes out through the antenna). Even if you add 30-40% more heat to it, it isn't significant. And a lot of folks prefer large and heavy instruments on their desks. Switch-mode supply gets really appealing when you are on-the-road but then you would use batteries anyway.

[floobydust]

What I saw with these linear power supplies is they got in the "race to the bottom".
Commonly made in Korea or china or Hong Kong, no safety approvals, over-rated components and then add in the cheap garbage electrolytic capacitors. Their quality is pretty low.

Same circuit, sometimes small variations. Most don't have a LM723, it's a few transistors. It's widely copied and scaled up from 3A to 50A even just my adding pass transistors etc. Sold under Samlex, Alan, PowerTech, Jaycar, Pyramid private labels as well.

Look at these Samlex pics- the power transformer windings aren't even tape wrapped, they are loose BUZZZZZZZZ. Caps took the piss. Heatsink so small and they get super hot even at 1/2 rated output. Let's push the 2N3055's to their limit.
It has EMI actually, from the rectifier diode switching hash and they don't have any disc caps to deal with RF except one at the output binding posts.

I see manufacturers changed over from linear to SMPS, not sure you can even buy these today due to the their weight. I guess it's not the 1980's.
Samlex America only offering SMPS now.

[radiolistener]

While it's certainly possible to have a switch-mode supply totally clean enough to power ham radios

I'm sure for 100% this is totally impossible due to physics limitations at least at adequate price, size and weight. Linear one always win in this comparison. You can create a high-quality switching power supply whose noise levels are satisfactory by shifting the spectrum of its main power to unused frequencies. However, even with such a power supply, you will still have switching noise across all frequency bands. The only question is whether you will consider their presence acceptable or not.

It is much more easy to not create noise, than trying to eliminate it. Adding even super cutting-edge high quality filter cannot help to filter noise completely, because there are a lot of paths for noise leakage. It's like adding an infection to water and then trying to purify it. Adding is easy, purifying is hard, and completely purifying it at a reasonable cost is almost impossible. Thus, an SMPS with low noise will be several orders of magnitude more complex, expensive, more heavy and larger in size (due to filtering and shielding) than a linear power supply, while still noticeably losing to the linear one in terms of noise levels.

This is why there is no SMPS in the world that does not produce EMI noise detectable by a good radio receiver. You can easily see noise from any SMPS with a good receiver. I didn't seen SMPS which noise is not visible on a receiver with pretty mediocre sensitivity, not to mention professional receivers whose sensitivity is comparable to the level of thermal noise.

[JoanBS]

The Riden RD6006 looks good but there's not much input voltage for it? I measure 22VDC at light load as my Samlex 13.8V PSU has. For charging say lead-acid batteries... how much headroom the RD6006 requires? There is the smaller RK6006 as well.

You could change over to a full-wave bridge rectifier (instead of the original center-tapped config), upgrade (voltage) on the filter caps and a fan, then it's looking pretty useful.

These 13.8V linear power supplies are cheap, made in Korea or china and also over-rated. The 20A is intermittent (TX) use I guess.
It's 30 minute max. at rated load then a cool down for 10-120 minutes at reduced load down from 20A to 8A.

I have a dumpster dive one, an inch from the garbage. The electrolytic caps all were bad and leaked. Replaced them and a few years later, I've never needed it, no use for a big heavy linear PSU taking up so much space.

I totally disagree. I have a power supply with similar characteristics, 13.8V and 30A, which I built about 40 years ago to power an amateur radio station. It has a large 500VA toroidal transformer and a simple circuit with only discrete transistors (no integrated regulator). It has current limiting (internally adjustable), and can withstand a short circuit indefinitely with 0 amps of consumption. The power section consists of a 2N3055 driving five 2N3771s in parallel, with static cooling and also forced air cooling, above a certain temperature.

I still use it every day, now in my workshop. I modified it not long ago and I can vary its voltage between 7.5 and 18V, and it's my first resource among a dozen other power supplies, to power all kinds of devices with power within its range. Its stability is unwavering, and it has never let me down in these 40 years. It's quiet, keeps as warm as it needs to be, consumes more power than an equivalent SMPS, and also weighs considerably more, but I hope it will be with me for many more years to come.

Switching power supplies and true amateur radio are simply incompatible.

[NiHaoMike]

I'm sure for 100% this is totally impossible due to physics limitations at least at adequate price, size and weight. Linear one always win in this comparison.

Small linear in parallel with a large switcher would be the cheapest overall and just as low noise as a linear during receive.

[Wallace Gasiewicz]

While I agree with radiolistener, there are Ham radios that come with their own PS that work quite well. Of course the spurious signals of the PS are outside of any critical frequency and the specific PS for specific Ham radios are typically very good for switchers. Also the radios themselves have amazing noise filters that probably were not as necessary in the past.     

If you wish to see the radio waves given off by a switcher, put an antenna near it and attach the antenna to your spectrum analyzer.     

Of course there are also all sorts of spurious signals running loose in today's radio environment.  From computers, wifi and ethernet and all the wall worts etc....so the additional noise from another switcher may not be significant ( if they are good switchers), although I have had a few switchers that made radio listening very troublesome. They also can grossly affect any RF measurements if you are doing any RF analysis while trying to fix something.