Choosing a linear power supply for rx-only

Status
Not open for further replies.

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
Just some general guidelines when choosing a linear power supply for long-term receive-only use ...

It simplifies trying to figure out all the "ICAS" ratings and whatnot, usually intended more for transmitting needs. Long term rx-only is very simple.

Don't exceed 25% of the total rated amperage of the supply. That means warm, not HOT. Warm is where you can leave your palm on the heatsink all day long - you'll feel it for sure, but you can leave your hand there. HOT means you simply can't leave your palm on the sink for more than about 10 seconds. Components will last longer without having to run ice-cold if you don't exceed 1/4 of the total rating.

This is even more important if the ambient temperature of the shack is hotter than 77F.

Example: My HF transceiver, used for rx-only, pulls about 1.4 amps max. With that, I use a 7-amp linear supply. Heatsinks are warm, but not hot. I tried the same with a smaller 4A rated supply, and I just exceeded 25% by a little bit. OUCH! Sure, the 4A supply is a little toasty, and is built to dissipate heat, but I want to extend the life of the components.

*** WARNING ***
If you feel your heatsink, DO NOT WEAR ANY RINGS doing this. Why? If you short the exposed power transistor case to the heatsink, you can damage the supply, especially if they don't have any other protection circuitry.

Considerations: If you have some really expensive gear on that supply, you may want to choose one that has "crowbar" protection. That usually means if something goes wrong and you lose regulation, it will trip the supply, and not let the unregulated 18 to 24v through to the radio.

What causes this? Massive overloading and/or very high ambient temps can cause this. Maybe not right after purchase, but later down the line. Power line spikes can also do this.

Protect the input to the supply: Some supplies incorporate a single transient MOV or other parts inside the supply. That's fine. But I STILL recommend an external unit, like a Tripp-Lite Isobar to do a much better job of that. Especially for the smaller supplies that don't incorporate a crowbar protection circuit.

It really is that simple - go big if you want to run ice-cold, or envision adding more gear to the supply later on. Or perhaps you need a feature, like a built-in crowbar that sometimes comes only with larger supplies.

Of course, if you want to take the whole thing to the next level, there's always this:

Astron Power Supply Index Page

Astron or not, it's a great resource.
 
Last edited by a moderator:

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
I hate to recommend, but the common 3 are the Astron, Tripp-Lite, and Pyramid.

Each has their detractors for different reasons. I only have experience with the Astron's, and lately the Tripp-Lite's. I do the MFJ thing, and open each up looking for obvious faults. Despite the lack of schematics, I like what I see inside the Tripp-Lites. Modern ones that is.

I have no experience with the Pyramids, but with all three manufacturers, you get everything from pie-in-the-sky reviews to "total junk" and threatening to cancel their subscription to linear-power-supply magazine. :)

For me, the most important thing, whether you plan to buy one, or already have one, is protect that input with a Tripp-Lite Isobar and keeping things cool by not running more than 25% of it's rated output for long-term rx-only. In this fashion, you may not absolutely *need* a crowbar. For a $3K rig - yeah, I'd want one. For other stuff - it's a judgement call.

But yeah, I've replaced many of the common LM723's, pass transistors, modified and upgraded the hamfest beaters. That can be a hobby unto itself. Replace all the 2N3055's with 2N3771's - sure. Does it bring anything to the table doing that with the smaller supplies? Not so much.

Thing is, these suckers are heavy - and some online retailers may not provide adequate shipping protection. Here, a brick-n-mortar purchase might be advised. :)
 

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
Common problem - immediately shorting the output with bad connections!

I've been guilty of this. In my excitement to get the supply online, instead of doing the right thing by using insulated lugs, I've just twisted wires, usually too big for the jacks themselves, and screwed them down too tightly.

POOF. Despite internal protection, the frog-hairs of the stranded wire shorted. Or the big solder-tinned cables crack the plastic connectors. Or short out when moving the supply back on the table.

For rx-only, if you aren't pulling a LOT of current, you can get by with banana-jacks connections. At least this gives you more than the typical 1/8th inch spacing between the jack and the case. And they rotate a bit when you move the supply around. Or just fall out instead of shorting.

Even though one should use totally secure lugs when transmitting to avoid voltage drops, for rx only that voltage drop is not as critical. I'm not saying one should be sloppy, but it would be better for rx use to use a good banana-jack connection than have a frog-hair short, or torque cracking your jacks with bigger wire.
 
Last edited by a moderator:

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
More common problems:

Might as well cover other things I've been guilty of too.

Don't use the power supply as your "master switch" to power up a rack full of stuff. Don't just hit it immediately with a big load of rx gear. Fire up the supply, and then turn the rest of the stuff on. Sometimes you can't help it with a power outage, but do what you can. You want the regulators up to full speed before put a load on it. No, you don't need to look at your watch. Just common sense.

Here's some anecdotal stuff - bordering on crazy :)

With a new supply, I'm not totally sure that the electrolytic caps are fully formed. To help make sure, I'll run the supply for 8 hours or so, and then turn it off for 5 minutes to allow them to discharge. Then fire up for another 8 hours or so. I'll do this a few times right after purchase for a few days. 8 hours is just an example, there is no definitive time period to do this.

However, the secondary benefit is to flesh out any problems during the warranty period to make sure it passes a simple on/off a few times. Is the power switch flaky? Was some other component on the edge and can't handle a simple on/off? Sometimes a problem can be hidden if you only turn on the supply and don't actually turn it off for a decade. It happens.
 

RFI-EMI-GUY

Member
Joined
Dec 22, 2013
Messages
6,859
Don't buy way more power supply than you need. As tempting as it may be to buy a bench supply that is variable to 50 volts you don't want to hook your radios to it and have someone turn the knob from 13.8 up to 48 volts. Get a fixed voltage supply for your radios.

When you buy a supply, measure the output with it unloaded. If it is not reasonably regulated like 14V for a 13.8 V supply, send it back. ICOM has linear power supplies built into their ICR-71, ICR-7000 and ICR-9000 receivers that REQUIRES a load to be in regulation. If you pull the DC jumper on the back of the radio, the voltage will run up to 18 or 19 volts! If you jam the jumper back on the panel lamps will fail (and potentially worse). This is a design "feature" due to a bootstrap resistor from the output of the regulator pass transistor back to the rectifier. It leaks enough current that the capacitor voltage rises. Otherwise they work fine. But you don't want a desk or bench supply that does that. Buy a power supply from a reputable manufacturer, one that has a crowbar circuit and you will be fine.
 
Joined
Jun 13, 2018
Messages
869
For ages I have used Astron power supplies- if I could number them professsionally they would be in the dozens. I also use them personally, and in all cases (knock on wood :))-- never a failure. Many of them have been in continuous service for years and years; from hot sticky tropics to the frigid lands of ice.

I can't sing these praises for other supplies. I have had some really beautiful failures that took out expensive equipment. Those memories linger, such that even though I trust the Astrons, I insist they all be the "M" models- ones with Voltage and Current meters-- those meters just make me feel better.

__________________________________________________

Yes, like Hertz says, watch the jewelry around these beasts !

I have a permanent scar about a finger when a ring got messed up with a 13 volt supply. I lost a nice gold bracelet to a similar slip.
Needless to say, I wear no jewelry around any equipment today.

Lauri :sneaky:
 

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
Well, I'd say that not exceeding 25% of the rated max output would be the *minimum*, and buying a little bit more than you need is the way to go, especially if the supply if very small to begin with. Why?

Cooler operating of course. This means less chance for component failure due to rapid heat aging. Made worse when ambient temps are above 77F. Thus less chance to lose regulation and fire that crowbar.

TIP: don't change the fuse with something larger than what was originally specified! The crowbar, if you have one, does it's job by shorting out the supply output (like throwing a crowbar across it), which is supposed to blow the fuse. If the fuse is too big, it doesn't blow, and even worse things happen.

A larger supply means beefier components, transformer, caps, rectifiers and the like. May handle incoming spikes better than a smaller unit would - again protecting the supply from over-volting (if it doesn't go the opposite direction and go low current-limiting from the failure).

My first thought here is before buying anything new, protect what you already have! The Tripp-Lite Isobar would be my FIRST purchase. New of course, not some oldy-moldy used online where the mov's have been worn out from decades of taking spike hits. With that, you lessen the risk of rectifier, LM723, or pass transistor failures. And possibly lessen the need for the crowbar hack. It certainly does better than the usual single MOV in an Astron.

More headroom for additional gear in the future. If you cut it close to the 25% recommendation now, you'll have no room for anything else and will want to get yet another supply if you value heat-reduction longevity.

Heh, crowbars - the bane of many amateurs who trip it with high rf in the shack. Or tickle the Astron's socketed LM723 regulator, and need to frequently replace it (hence the socket) as a band-aid measure instead of fixing the REAL problem. Oxidation in the socket makes it more vulnerable in the first place, so a catch-22 begins to happen. :)

Astrons are nice, but in the end, these are still "consumer grade" 1970's tech supplies. Use them to power radios, NOT halogen bulbs, makeshift cornball battery tricklers, and some common sense, and the average listener will go far.
 
Last edited by a moderator:

majoco

Stirrer
Joined
Dec 25, 2008
Messages
4,278
Location
New Zealand
Don't use the power supply as your "master switch" to power up a rack full of stuff.

Excellent advice from Hertzian for two reasons. If your power supply has a big transformer for the main output and a smaller one to run the regulation/voltage/current control then it's possible that the main output will be unregulated for a half-cycle or so and go to the maximum before being brought under control - if it's a 48v max supply then I hate to think what might happen in those few milliseconds.

Secondly, if you have a lot of equipment hanging on the output then there's likely to be a lot of discharged capacitors that need filling up before anything comes on line - the surge current could be very high and the overcurrent protection in the power supply hasn't woken up yet - the output transistors usually fail in a shorted condition.... oh dear.... :(

Some years ago I was an instructor of apprentices in a 'hands-on' environment and we made a good quality power supplies of about 20 volts at a couple of amps for each bench of two trainees as a "design and construct" exercise. (The original design was from Lyndsey Hood in the "Wireless World magazine.) It didn't take long for the boys (and girls) to demonstrate that they could blow up anything. Charging flat batteries was the main culprit. They needed a few modifications to make them bomb-proof!
 

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
Agree with you there Majoco!

Don't even get me started on how unsuitable these things are for battery charging. Yeah, they will - just like I can wash my clothes with a bar of soap and a rock at the curbside. :)

Why I STILL use 1970's tech linear power supplies for receiving:

Sure, I've tamed older SMPS supplies with ferrites and whatnot, and even been very pleased with more modern ones with very little rfi, frequency offsets and the like.

BUT, since I'm using low-signal output antennas like small loops with high s/n ratios, I can still hear things that guys with an S7 noise floor might not and proclaim in a review to be noise-free. Heh, not for me.

I can rotate the loop and do a bunch of other things, like run off a battery as a test - or rotate the frequency offset, but that just gets to be a pain in the long run. So ROCK-ON, you linear power supply users.
 

WA8ZTZ

Member
Joined
Feb 23, 2014
Messages
971
Back in my apprentice electrician days eons ago, the journeyman (and also a ham) who was my patient mentor told me that he
always bought his electrical gear (and ham gear) "by the pound"... IMHO, especially good advice when buying a power supply.

btw, if using a power supply for charging or floating a battery (generally not recommended), make sure to have fuse protection at the battery in event of a short circuit with the leads or if the crowbar were to fire.
 

RFI-EMI-GUY

Member
Joined
Dec 22, 2013
Messages
6,859
Back in my apprentice electrician days eons ago, the journeyman (and also a ham) who was my patient mentor told me that he
always bought his electrical gear (and ham gear) "by the pound"... IMHO, especially good advice when buying a power supply.

btw, if using a power supply for charging or floating a battery (generally not recommended), make sure to have fuse protection at the battery in event of a short circuit with the leads or if the crowbar were to fire.
I charged a car battery using a linear supply one time and after turning off the power, the pass transistor apparently got reverse biased and failed shorted .

Sent from my SM-T350 using Tapatalk
 

rja1

Member
Premium Subscriber
Joined
Dec 14, 2004
Messages
156
Location
Enfield, NH
I've had an ASTRON 35 Amp RS-35M power supply running 24-7 / 365 since 1991. Rock solid reliable. Not even a hint of RFI, unlike most switching power supplies. Only downside......it's heavy, lots of iron needed for 60 Hz transformer.....but it's not like you carry it around with you. That's my experience, YMMV.

Bob
N2OAM
 

Ubbe

Member
Joined
Sep 8, 2006
Messages
8,944
Location
Stockholm, Sweden
The only thing that could do RFI in a linear PSU are the rectifying diods when they switch on and off. The good PSU's have cheramic capacitors over the diods and the bad ones have holes and marking on the circuit board for capacitors but where never installed to save time and money.

/Ubbe
 

majoco

Stirrer
Joined
Dec 25, 2008
Messages
4,278
Location
New Zealand
I charged a car battery using a linear supply one time and after turning off the power, the pass transistor apparently got reverse biased and failed shorted .

Apparently? Certainly! The same can happen if your load has a large capacitor in it - protect the pass transistor with a normally reversed diode across it and a diode in series with the output capable of the full output current plus surge - or better still a relay that disconnects the output if the mains power fails.

Better still, use a properly designed battery charger!
 
Last edited:

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
If you don't make your own, I think Astron has some models with built-in battery charger circuits. Still, I leave charging to actual chargers...

Ubbe - good point. I usually put a 125v or better .01 to .1 uF cap across the positive to ground of the rectifier bridge. Sometimes a bad solder joint on one of the diodes exacerbates things, and that quiets them down immediately. All part of my "MFJ Inspection" when I get a new one.

Tip: Don't Baby that power switch - flick it!

Sometimes that "thuuunnngggg" sound one hears as you turn on a supply, and it surges initially to charge a fully discharged main filter capacitor, can be scary to the unwary and they try to be "gentle" to the rocker switch in an attempt to make that power-on sound a little quieter.

The problem with babying a nice feeling rocker, or moving a slide-switch slowly is that with all the current trying to pass at power on, the slow switch-on is hard on the switch as it tries to pass huge currents with a dinky mechanical connection before it makes full contact. FLICK that thing, dont massage it into place.

Sometimes being too slow on the switch will blow the fuse. Do NOT go above manufacturer's ratings - that will defeat the purpose of the crowbar. I've also seen some duly put a properly rated fuse in there, but they used a fast-blow instead of a slow-blow.

And back to those rectifiers - again one of the reasons why I cycle a new supply a few times over the course of several days before the warranty expires, is to flesh out any problems with the rectifiers from simple power-ups. But don't be crazy and just sit there flicking the supply on and off. Give it a little time.

Initial power-on surge stresses the rectifier diodes a little bit - it they are marginal to begin with, I want to know up front. OR, if a bad solder-joint is the culprit, I want to know NOW, not 10 years later if it never gets cycled.

Some put varistors inline to dampen that a bit at power on - repeaterbuilder has all the more detailed info....
 

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
A note about Tripp-Lite power supplies ..

I have recent (2018 manufactured) PR 4.5a and also a 7B 7A unit. And I have to laugh at certain reviews.

I use the smaller 4.5A for running receivers that draw no more than 1 amp continuous since I like to run cool. Likewise, with the 7b, I run larger receivers that draw about 1.5 to 1.75a continuous - again since I like to run cool. Well, warm, but not HOT to the touch.

They seem to be well made - even with ROHS soldering. Unlike Astron's (which I also use), the regulator board is not screwed onto the main filter cap, but resides on it's own sepately mounted board. Parts are very common to all the rest, that is your usual LM723 in the larger units, TIP drivers for the 2N3055 / 2N3771 NPN power transistors, electrolytic filtering caps rated at 105C temp, a smattering of bypass / filter dinky .01 to .001 caps, balancing resistors made of metal oxide / matched resistors etc etc.

The case itself is very well painted inside and out - so much so that you might want to run one in the garage for a day or so, unless you like the smell of fresh paint. Not toxic, but the ones I received were obviously packaged up tight right after manufacturing sealed in a bag and box with not much time to air out. After a day or two it's gone.

The funniest thing is the whining about not getting a schematic with it. I have to agree with Tripp-Lite here - if you don't know what you are doing without a schematic for generic power supply parts, then for your own safety, take it to someone who doesn't need one. :)

Oh, and to help the guy who just shotguns parts, common part numbers and designators are printed on the board itself. Perhaps when doing an MFJ inspection, one will want to look around and take inventory. Much of the stuff is easily available.

I dunno' - seems like most of the knee-jerk reactions about Tripp-Lite supplies are unfounded, and sound like marketing FUD. At least to me.
 

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
Common problem: using your supply as a ground plane for mag mounts that transmit

Not so much an issue for swl's, but can be one reason why Astron supplies have socketed LM723 or other regulators. Or a warning for the guy running high power into an indoor antenna with a poor ground. Tickling that regulator with rf is not a good idea, but socketing the regulator makes it an easy parts swap for amateurs. I've done it because of that. The trick is to fix the real problem of rf in the shack or on the cables. It's not a bug, its a feature! :)

Socketed pass transistors on Astrons? On something that goes through a lot of thermal expansion and contraction? What are you thinking? Yep, abused supplies that get run to the limit or beyond have easy to service parts swap / checking of the pass transistors with sockets. The problem comes from that junk box replacement part that have a large coating of solder on the pins, jamming the sockets wider when pressed in, and replaced yet again later - in addition to thermal expansion contraction. Sloppy cleaning techniques that press dirt etc through the mica insulators, and not replacing the thermal compound with new is another. (although it can be dry, the big issue is breaking the initial microscopic seal and not applying a new one).

I'm not bagging on Astron - I've had a few that performed extremely well for a looong time. Marketed mainly to amateurs, where the environment can be rough, I can see where socketing things that shouldn't be socketed makes sense to get back on the air - even if it is only band-aiding the real problem for awhile. Stuff that an swl'er doesn't subject the supply to.

Moral: treat your supply right in the first place. Take the hint from Astron with a single MOV transient absorber on the front-end that you might want to add additional front end protection (a GOOD protected power strip) regardless of brand. Those strips are easily pilfered at repeater sites, so modification of the supply is always a good idea, but for home use, you can do even better with a high quality strip.
 
Last edited by a moderator:

cmdrwill

Member
Joined
Mar 30, 2005
Messages
3,984
Location
So Cali
That IS why they call them ASSTRON's. Not the best design.

And Herrtizin did not mention piss poor connection to the pass transistors via the mounting hardware. Add internal star tooth lock washers on the collector screws.
 

nanZor

Active Member
Joined
May 28, 2009
Messages
2,807
Yep - and a star lock washer for the case ground if it uses that. Not just for mechanical integrity, but to "bite" into the metal for a solid contact.

I don't mean to bag on Astron - I've had many do yeoman service over the years. Most of the biggest problems come from overloading and treating them badly. Like any other manufacturer. Some are better than others, but I see waaay too much marketing hype from users and prospective users that have an ulterior motive.
 
Status
Not open for further replies.
Top