Strictly Linear PSU's

Do you enjoy making your own equipment or antennas? Discuss construction and design in here.
Post Reply
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

Finished ! All good so far, see the meters for unloaded and loaded with 150 Watts of halogen lighting. Transformer ran at maximum of 76°C transistors at 52° C and the rectifier at 48°C run this load constant for 90 minutes. On reflection I could have easily achieved 20 Amps with two more transistors. It's not very economical on primary power, so it's akin to a coal fired boiler. For the 10 - 12 Amps I aimed for, I could have wound the MOT secondary with 16 AWG, I used 14 AWG which could handle twice the output current I ended up with. In conclusion, and given a 900 to 1000 Watt salvaged MOT, with regard to output current, the sky's the limit. If you attempt a project like this, do the homework first, and if your at all unsure, go buy a retail power supply. It's been good fun, not sure when or what's next, see what the future brings.

PS. The unloaded voltage shows front post, and rear sockets voltage. Reason for two meters reading 12.59 Volts.
Second meter shot shows current on the red meter, and voltage on the yellow meter.
You do not have the required permissions to view the files attached to this post.
CB call. Shipwreck.
oh5nxo
Regular
Regular
Posts: 40
Joined: 24 Apr 2017, 18:41
Call Sign: oh5nxo
Location: kp30xr

Re: Strictly Linear PSU's

Post by oh5nxo »

How efficient is it? My old clunker wants 120w AC at 72% power factor when rig needs 60w to create 30w of RF. Depressing.
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

oh5nxo wrote:How efficient is it? My old clunker wants 120w AC at 72% power factor when rig needs 60w to create 30w of RF. Depressing.
In terms of primary to secondary current it's a monster, when I loaded the AC with 300 watts it dropped less than a Volts, from memory about 0.4 Volts. So it doesn't droop under very heavy loads, even when it's rectified and regulated. I've a 300VA maplins transformer, twin 120V primarys, twin secondarys, even in parallel secondarys drops a lot under load, couple of Volts at least. The MOT primary runs heavy from the start (no load) this is due to less turns count primary. At start up its drawing 2.9 Amps, this is assume is wastefull of power, cooling is an absolute must in operation. So it's as efficient as a coal fired boiler...
Those green sustainable power users would not agree with its efficiency. It's archives a means to an end, I should have used way more transistors to get the more current that's easily avaliable. It was a model really, proof of concept if you like.

What people must understand is once it's salvaged from a microwave, the high voltage secodary is cut away, that's to allow for a heavy gauge low turns count low voltage secodary winding. It's possible to add turns count to the primary to improve efficiency, but then you need extra turns count secodary to achieve the the output voltage you need. Space on the transformers frame is at a minimum, as it was originally designed for high voltage many turns thin wire secondary. Under no circumstances should anyone power it up with the high voltage winding on it. For those interested the Web, YouTube, and many sites have tutorials on converting them for low voltage use. So if it's for a DX session, and your powering that high wattage amplifier, it will do that easily. It's not efficient to use its 24/7 in terms of your power consumption from such a power supply. I'm not a physics guy, but it's not greatly economical in use, but for shear grunt in terms of secondary current, it's there by the bucket load. I would suggest using at least 4 x power transistors, and possibly more to make use of that secondary current. Once again, unless your compleatly at home with electronics, you should go and purchase a retail power supply. Many people have made use of salvage MOT's for different applications, in this instance it was for low voltage high current. I'm pleased with the outcome, if it only ever sees occasional use. Loads of topics on electronic forums, and many other sites. But always play safely.
CB call. Shipwreck.
oh5nxo
Regular
Regular
Posts: 40
Joined: 24 Apr 2017, 18:41
Call Sign: oh5nxo
Location: kp30xr

Re: Strictly Linear PSU's

Post by oh5nxo »

Most of the year, extra heating is quite useful. I assume it holds for UK too.

I might have also noted in the last post, that the same 550VA clunker without load, takes 0.25A, 11VA 17%, only 2W.
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

Not sure on the MOT VA rating, i presume its going to be high. Certainly three figures I'd guess, its not efficient from a consumer point of view. I'm lead to believe the primary idle current is some what wattless, kind of standing power, either way its wasted heat, and hence the need for cooling. The low turns count primary, gives a very rigid secondary, no volts drop, or tiny volts drop even under haevy loads. The 90 minutes i ran it for did warm up the kitchen lol, i could see no real dent in electrical consumption on the house meter from that run time. Just because its waste full of power primary, i personally wouldn't use it as a main RX radio psu. Although I'd be happy to drive a 150 watt amplifier with it, well, for as long as a DX session lasted. My MOT was from our knackered 700 watt microwave oven, the magnatron kicked the bucket. My advise would be try to find a scraped 900 - 1000 watt unit to repurpose. And go with MJ11015 if your circuit requires pnp transistors, or MJ11016 npn transistors, these are high power high gain darlingtons with maximum collector currents of 30 Amps. The high gain hfe 1000 means any voltage drop under load is minimal. I was running 150 watts drawing 12 Amps, the transistors never got above 52°C very good i think. So using 4 x or even 6 x transistors would be more like it. You could use any linear regulator circuit for the voltage control, LM317, 78XX series, LM723, etc. There are others also. With regards to the transformer, i removed the shunts, i wanted the maximum current. You could leave them in, it would probably run cooler, although there might be a reduction on current. With an LM317 or LM723 a voltage of 13.80 Volts would be achievable. Bearing in mind your turns count on the secondary and its final AC voltage. In operation on a hard surfaces i can notice a moderate transformer hum (Audiable) On the carpet the hum wasn't noticeable this has no baring on output electrical noise. The output would still be clean. A project worthy of a go, given saftey practices are followed.
Last edited by sureshot on 15 Jul 2017, 12:00, edited 1 time in total.
CB call. Shipwreck.
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

Just another couple of tips on preparing the transformer, worth a mention. Going back a couple of years now, i did reclaim and rewind another MOT. In that one i did seperate the EI sections of the core, on ressembly i measured a higher primary current draw than before i cut the core. I know its a pain winding a closed core, but in my opinion it preserved the frames magnetic circuit better. And as your unlikley to want more than 20 Volts secondary, it doesn't take to long. I lined my core with thin stiff card to facilitate winding isolation, and my wire choice was high current silicon wire. I used heatshrink on 10 meteres of it, i used just under 5 meters. String pushed through the heatshrink with a steel rod got the job done. The heatshrink probably would not be necessary if the wire quality is very good, mine was thin on insulation in places. Teflon tape rap is also recommend for insulation of winding wire.

As for removal of the old high voltage secondary , i carefully cut down one side of the windings with a hacksaw, taking care not to damage the primary windings. Then i decided to drill through the copper sections with an 8mm drill bit, it was messy, but got it done. You could just cut one side of the old secondary and use a drift to tap it out. The new windings must be tight on the core, and even over each row. I used 14 AWG, but i could have used 16AWG and still achieved a higher secondary voltage, i got 14.70 Volts AC secondary voltage. No more than three rows would fit on my transformer core. 16AWG would have given me more turn and a higher voltage in reflection. This is why I'd suggest looking for a scrapped 900 - 1000 watt unit, for greater turns count on the secondary. As with any psu projects always play safe, use an RCD for any transformer testing, and always ground the transformer frame. Once again if your at all unsure, go out and purchase a retail power supply. Any psu project requires strict respect for saftey.
CB call. Shipwreck.
User avatar
Stronty
Veteran
Veteran
Posts: 2795
Joined: 24 Sep 2007, 17:48
Location: Up North

Re: Strictly Linear PSU's

Post by Stronty »

@Sureshot Just wanted to say I have enjoyed reading and seeing your contributions to this thread. Very interesting all round. thanks for taking the time to log it all here.
Can't help but love an Adams.

Image
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

Stronty wrote:@Sureshot Just wanted to say I have enjoyed reading and seeing your contributions to this thread. Very interesting all round. thanks for taking the time to log it all here.
Thanks Stronty, if a handful of people can use some ideas from it, then it was worthwhile. It might be a while before I get round to another project, it was tough finishing this one (rare anemia problem not getting better) But I try block the negative stuff out as much as possible. Anyways, I enjoy it, I doubt at the moment it will see much use. My biggest amp is only a KL 200.
I'd like something a bit bigger :D but maybe in the future. I'm no electronics expert, but a hobbyist, frequenting a few electronics forums, there help has been amazing. If I can get on top of some health issues... I will certainly be up for another project. I'm glad you liked the thread. :)
CB call. Shipwreck.
User avatar
cb4ever104
Legend
Legend
Posts: 6551
Joined: 28 Jul 2011, 21:26
Location: Alicante , España

Re: Strictly Linear PSU's

Post by cb4ever104 »

Stronty wrote:@Sureshot Just wanted to say I have enjoyed reading and seeing your contributions to this thread. Very interesting all round. thanks for taking the time to log it all here.
+1 :thumbup: :thumbup:

Could do with more threads like these . Reminds me of the good ol' days with RD250 ........
30TM060
Paul aka "Sammy". (108) PY60 Old Paisley DX Group (RIP Rab Markie)
Ex GM0 , now EA5 .
Handle "Fortune Hunter" back in "the day"
Radios ....? Anything that says "TAIWAN R.O.C" on the back.
User avatar
Auldgeek
Veteran
Veteran
Posts: 4189
Joined: 05 May 2014, 09:18
Call Sign: GM0BRJ
Location: Lanarkshire, Scotland

Re: Strictly Linear PSU's

Post by Auldgeek »

+1 from me too. Excellent work. :thumbup:
Auldgeek - Drew

Winner of IBTL Autumn 23 Edition
I've changed radios so many times, I've forgotten what I have :think:
User avatar
Buick Mackane
Moderator
Moderator
Posts: 9560
Joined: 21 Aug 2012, 17:30
Location: A citizen of legoland

Re: Strictly Linear PSU's

Post by Buick Mackane »

Another thumbs up from me dave :thumbup: Excellent work and a gold mine of info for anyone thinking of building a PSU.
I Am the great cornholio are you threatening me ?
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

Cheers guys, hope a few can get something from the thread. Converting a MOT's not for everyone, but if a few go on to build a linear power supply from a purchased transformer, then it was worth it.

I've got a few server smps under the cupboard for future projects in the other thread. I'm already looking at other components now, lol. Hopefully something to come in the near future.
CB call. Shipwreck.
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

Been a while since I posted in the thread. I've been saving a couple of old pc cpu heatsinks for a future project. I think they make great size for power dissipation in a relatively small footprint.

Wanting to try another couple of power transistors, see how they perform. These heatsinks I'm using the TIP36C and the TIP147 both PNP.
The TIP36C has the higher collector maximum current, and the TIP147 has the more gain.

Both are fairly easy to come by, although the TIP36C is not quit as easy to source. I avoid eBay for semiconductors, as I've been bitten before by fakes.
The two heatsink picture used selftapping screw of 3.5mm cutting a 3.2mm hole first. It's quick dirty but works suppressingly well. I didn't think they would cut into aluminum and copper very well. But they did, and bite just fine.

So I'd had a couple of Intel 775 stock coolers, but they offer a shade less space for silicone that the 775 heatsink in the picture. That one in the picture is off an Acer Aspire pc, much more substantial than the stock Intel heatsinks. This time I thought I'd try tapping M3 threads, and try my best not to snap a tap. That went well, but took longer. Managed four holes and no snapped tap.

We've all been there at some time with that snapped stud in an engine block. Shock horror, but I really don't think extracting an M3 broken tap an easy task. Any way a couple of units for a future project maybe. Anything up to 100 watts on these heatsinks isn't unreasonable, even more so with duty cycle that radio use affords most power supplies.

Anyone else made anything in the way of a linear power supply ? Post up if you have. Anyway until I have more.. :)
You do not have the required permissions to view the files attached to this post.
CB call. Shipwreck.
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

Thought I'd put this up, might be of interest to someone out there. I've been collecting old graphics card heatsink for a while now. Not sure why, must be my hoarding instinct :lol: So a few weeks back I purchased a second hand graphics card off eBay, for use in a server cpu desktop combo I'm trying out. Sadly one morning there was a distinct crack as something hit the floor after the postman pushed it through the letter box.

My purchase was wrecked, heatsink ripped off the pcb, repair not possible :cry: I salvaged the heatsink, and that was about it. For a while I've been thinking of using these old heatsinks in a modest size power supply board. The attraction is the low profile that might be possible with such slim cooling hardware. With fan cooling through the case, I wrecon 3 - 5 Amps should be possible. 50% duty cycle at the higher end, and continuous at the lower end. It's far from a huge power supply, but adequate for an 80 channel radio or scanners. Maybe even a small amplifier.

I've taken a couple of pictures to give some idea of what I'm on about. The stripboard I used is FR4 rather than the more common phenolic boards, difference being rigidity. As the spring pins excert a force on the board when installed. You have to centre the components well to achieve best fit and contact with the heatsink. As components like through hole tend to be a bit thicker, a few coils of the the springs needs clipping off. As does the captive shell retainer used when there installed on graphics cards. But the result is a low profile perfect fitting heatsink.

Anyway thought it worth a mention, I've not got round to soldering yet. But will post the finished power supply in the near future hopefully. You decide if it's worth the effort, I think it does, especially if you like tinkering. :)
You do not have the required permissions to view the files attached to this post.
Last edited by sureshot on 01 Feb 2018, 21:24, edited 2 times in total.
CB call. Shipwreck.
User avatar
sureshot
Veteran
Veteran
Posts: 2689
Joined: 24 Jan 2012, 21:26
Call Sign: 26TM413
Location: South East Coast UK.

Re: Strictly Linear PSU's

Post by sureshot »

My mistake, three images above.
CB call. Shipwreck.
Post Reply