I'm getting REALLY ANNOYED with the misinformation Jens is spreading on Twitter in response to this video. So here's a few points: Jens claims Meanwell's are bad, but original Commodore PSU's are good. What he doesn't mention is that for most of the PSU's Commodore shipped - They are EXACTLY THE SAME TOPOLOGY. The only difference is the Meanwell's use more modern controller IC's, and probably a higher switching frequency. They also implement undervoltage lockout, and overvoltage shutoff - which none of the original C= ones did. He claims that the Meanwell's do not compensate for voltage drop in the cable. That's correct... but NEITHER do the original C= PSU's. If they did, they would have an extra wire on the +5V line for remote sense. One or two of them even have a preset pot inside for adjusting the +5V output - exactly the same as the RT-50B and RT-65B do. Jens's design doesn't have remote sense either. He is "faking it" by measuring the current draw on the 5V line *at the PSU* and then boosting the voltage when more current is drawn. That works fine assuming he has characterised the voltage drop of the cable on the PSU, but here's the thing - the cable drops negligible voltage. Most of the loss is happening in the pins of the PSU connector. The only thing that helps with this, is contact cleaner. He claims that the regulation feedback from the Meanwell's is not accurate because it's not on the +5V line - IT IS on the +5V line. He is thinking of ATX PSU's which quite often only take feedback from the +3.3V and/or +12V rails. The Meanwell's also use the same TL431 voltage reference IC that countless PSU's, including all of the "off line" switchmode supplies that Commodore shipped. He's claimed that Jan's PSU connecting the shield ground to mains ground is incorrect. Guess what - that's exactly what the original Commodore PSU's do as well... want to confirm that ? Get a multimeter in continuity mode, put one probe on the outer shield of the Amiga plug, and another on the Earth pin of the mains plug. It will beep and show practically 0 ohms reading. Finally, he says the ripple and noise of the Meanwell's is not good enough for Amiga use. This is demonstrably untrue... recently "Retro Theory" has done just this on his Twitter account. They meet or exceed the ripple and noise produced by the original Commodore supplies. He claims people who have done this test have only done so off-load - Retro Theory has clearly done it on-load with an Amiga attached, and I have done it by loading the power supply with high wattage resistors. Retro Theory's test is better as it shows the response with a dynamic load. (I say "Commodore" supplies - but Commodore had nothing to do with them, they are all subcontracted. All Commodore specified was required voltages and minimum required currents. How the subcontractors did that was up to them. There were several vendors - off the top of my head: Speedy-Tech, ISMET, DCD, DVE, Phi-Hong. None of the designs were anything special, and a few of them were dreadful!)
A past sin revisited and rectified. :) That new case does look so much better, and I see no problem with a mostly new unit not masquerading as an original part. If I ever decide to take the plunge and build a PSU, I'm definitely revisiting your videos first. Thank you for sharing your knowledge and experience!
very nice build! May I suggest that you check that earth connection actually made it to the metal frames themselves? Unless I'm mistaken you checked continuity to the screws but there is a small chance that the metal (painted) cases are actually not making contact with them. Unlikely but better safe than sorry! :) You'll have to scrape the paint on a hidden corner to check. Thanks again for the nice video!
@@jaycee1980 exactly. I got a Chinese pre-heater board from Ebay and it came unearthed. The wires were there but were attached to painted surfaces. I sanded the coating down and, as you say, added a star washer to bite into the metal.
The icomp psu was my only option to run my a1200 ppc with bvision, SubwayUSB ect stable. Original C= psu began to shut off my system, atx psu damaged the Subway Controller, the system refused to start with another meanwell-based psu (but worked with my unexpanded a500) A neverending story. For me the icomp psu solved my problems. For this built: looks very nice👌 Great work!
Yes, the iComp PSU is a magnificent piece of work that combines 100% compliance with original Commodore specs and still delivers more power. I can imagine that it is the go-to solution for some more elaborate setups indeed. The Meanwell is a simple solution that works for most generic setups (and my slightly expanded Amigas). I have high respect for Jens' work on the iComp PSU, I actually chatted about it with him a while back and he explained the specs he had to work with to me. Excellent work by him and his team indeed even if it seems a bit over engineered (but for good reasons apparently). :)
If only those Bahar enclosures were available here in the USA. However, I did make my own Amiga power supply with a plastic enclosure and the same Mean Well RT-65B. I didn't have an original PSU to start with, so I used a square DIN connector that I bought off of Ebay, some vehicle trailer light wiring for the DC side, a normal toggle switch from the local auto parts store, and a power cord from a dead power strip for the AC side. It worked perfectly, thanks to Jan for his videos!
This looks great now! I also have a project box, having the same switch and a very similar Meanwell power supply to yours and it's working very well for my own Amiga.
Great follow-on to the prior video! I bought an inexpensive manual (not electric) nibbler a while back and haven't used a Dremel since for this kind of work :)
I'm using the same PSU for my A1200. I had a longer lead to the Amiga and practically no load on the 12V which floated the 12V too high since I had to adjust 5V a bit higher to compensate for losses so that I still have 5,0 volts on the computer board. Fixed it by jumping the +12V out via a beefy diode to the NC terminal, added a load resistor from NC to COM and took the +12V out from the lowered voltage NC terminal. I expect the 12V goes up more and more as you load the 5V with more amps and the switcher does more work.
Oh, good point. I’m going to keep an eye on that. According to the datasheet, none of the rails needs a minimum load but I guess that’s not the whole truth in this case. Good to know.
I built a new computer a year ago. I needed a dremel and a drill for it. I bought dual 360mm radiators. The top one didn't fit because the hole was too small. So I dremeled it out. But then the water pump had no place to attach so I drilled into the hard drive cage. It turned out really well. At the same time I drilled into my C64 so I could finally have a proper mount for my VIC-II² switch. Also turned out well. Surprisingly the plastic on my C64 is in amazing condition. It's very soft not brittle. The dremeling of the case though, that took 2 cutting bits. It's a thick steel case. If I needed to do anymore I would have used a 3rd as the 2nd one was pretty dead. lol Also I didn't have a stepper drill bit so I just started small and kept going bigger and bigger until everything fit. lol
I think my A1200 PSU is on its way out, it keeps freezing until it warms up for a few minutes. So, I think I'll be doing this build soon. Not sure to re-use the PSU case or buy/print one. Oh and a couple of step drills are among my favourite tools - I can't believe I didn't get them sooner. Great work Jan.
Your build skills have for sure improved over the years, I'd say your earlier builds look like "Bastelwaren", this one definately has a nice "German" look and looks like it came out of a factory.
7:28 2A would be the max input @100V (Japan) - fusing a 65W supply with 3,15A for 230V (=725VA) seems a little off, although the data sheet mentions an inrush current of 50A at cold start.
The internal mains fuse in the Meanwell PSU is a 4AT polyfuse, I guess I’m not off by that much? I’m not an expert though, obviously. The extra fuses on the IEC socket are just kind of an extra layer of protection in case some of my wiring shorts out catastrophically. :D
The inrush current is what flows when you turn it on "cold".. but the current flow lasts for a couple of milliseconds at most. This is why a T-type (time delay) fuse is used, it will hold for some seconds before blowing. In the case of a switch mode power supply, the inrush current is caused by the large 400V capacitor charging from empty. Certain kids of transformer, notably toroidal transformers, have an inrush current too.
@@jaycee1980 So the cap is there to make the inrush a non-issue and to provide lower noise and more constant current down the road ? I'm not an engineer so bare with me. :)
@@VincentGroenewold No, the capacitor is what causes the inrush. In a switch mode power supply, the mains voltage gets rectified and smoothed into a high voltage DC supply which feeds the primary side of the transformer. When this capacitor is empty, it draws a large gulp of current on power up as it charges. edit: to make this clearer, since as people will say "But transformers dont run on DC" - what happens is the mains voltage (which is 50-60Hz AC) is smoothed into ~380V DC, which is then "chopped" by a MOSFET into high frequency AC (usually anything from 50-500KHz), which drives the transformer. The reason is that the higher frequency means the transformer core can be much smaller than the equivalent core needed for 50-60Hz. This is why SMPS supplies are smaller and more efficient.
@@VincentGroenewold The capacitor is the main thing that cause the inrush current when started "cold" (it has to be charged), but it's necessary for proper function. The input capacitor needs to scale with the max output power (it needs to be able to provide power for a set amount of time) and if it's this is not managed in some way the inrush current scales linearly with the capacitor size. If you look at bigger PSUs they often have something to reduce the inrush current, like a NTC resistor or an inductor, and often on higher-end models a relay is used to bypass it once the in-rush has pased to avoid the efficiency loss (it's only desirable for that split second during cold start). The inrush current is also often very different on 110V and 230V - it's tends to be a much bigger problem with 230V since naive/simple constructions (like this PSU) often end up with 3-4 times larger inrush current on 230V compared to 110V, the RT-65 series doesn't even bother saying what the inrush current is on 110V. The RT-65 series says that typical inrush current at 230V is 50A which is a significant inrush current for a PSU this small, as a comparison I tend to want 600-850W PSUs to not go above a similar 50A-60A inrush current! and those have WAY bigger input capacitors (~10x) so they'd probably draw 400A+ without "in-rush protection" which would likely immediately trip your breaker even with nothing else connected! IIRC 60A or 70A inrush current is where Tom's Hardware start mentioning "high inrush current on 230V" on the downsides (IE warnings) section for a PSU test. Managing inrush current can be important when you have many *or* powerful PSUs, if your breaker immediately trips when trying to get power back either something broke or you have an inrush current issue. If you can get all of it working using the "unplugging dance", IE unplug you "enough" that you can reset the breaker and then plug stuff back in one at a time to spread out the "spikes". If that worked the the issue is that the combined inrush current is too large. I suspect there's no inrush protection at all in this unit which they only get away with because it's "only" 65W (so it has small capacitors) and is cheap. The 80mV ripple on 5V isn't fantastic either but it's the max possible value (so likely better or much better) and is likely fine for anything other than high-end audio anyway! Certainly I wouldn't expect any issues with the hardware he's planning to power with it. The 50A inrush current per RT-65 does mean it's probably not the best PSU to use if you want to say use a bunch to power a whole fleet of machine for an exhibit for example, but it's unlikely to be an issue in most other scenarios.
Nice, though I would probably have put some rubber (or plastic) pad/feet between the PSU and the case for vibration dampening and avoiding that "tin" sound if you tap the bottom. Not really important, just a very minor Quality of Life tweak you can do at any time.
Finally this PSU project came out really nice! I also like this case, looks very decent and solid.👌 Maybe I'm going to build a double PSU like this for my Atari 520 STM plus SF 314 Floppy, we'll see. 😉
Looks like no stranded wire for the mains switch? If so that’s definitely no recommendation from me. Unfortunately it’s hard to tell since assembly is always fast forward.
@@JanBeta Good decision, like I said, it’s hard to tell in fast forward. My Vids are usually very boring since soldering takes forever and is not entertaining but on the other hand I’m too lazy for editing…
Had for a long time a ATX PSU with a nosy fan. But now for 2 years a iComp CA-PSU. Have four of them for my A500 and my three A1200’s. Have to many money in my A500 and A1200’s. Do the safe way and not the dumb way such as my old ATX PSU.
I fitted a similar Meanwell PSU inside my Commodore 1541C drive last year, the linear supply was just insanely heavy and ran hot when idling (mostly trying to get it to work with VICE with a driver that windows refused to install), with the new one, it's light and runs cool, still have the original PSU stashed away though, just in case I ever needed to refit it... :D
I used the bare metal ones because the black ones are usually painted and don't make contact as well. But it probably wouldn't matter in this case (because of the washers etc) and would indeed make for a speaker look. :)
Should use some "star" washers underneath the screws securing the power supply. The washers reduce the screws' loosening from vibration and cut into the paint for an improved connection to the metal.
Great video, Jan! I'm thinking of building a universal psu for my retro computer stuff. Basically one psu with all the appropriate voltages and a couple of generic utlets, and then different power cables for the computers. :)
@@JanBeta Jens Schoenfeld says this, but it's because he wants you to buy his product. The Meanwell power supplies are better in every way than the original power supplies which shipped with the Amiga in the first place.
@@djbase_ As an electronics engineer, I can tell you right now that you are wrong. There is nothing wrong with the output of the Meanwell power supplies that would cause any issue with an Amiga.
@@VincentGroenewold He is repeating the rubbish spouted by Jens. Jens claims that the ripple and noise of the Meanwell's does not meet Commodores spec, and points to an internal memo released by Commodore to prove it. However, in fact the Meanwells have superior specs to the PSU's that Commodore *actually* shipped... so it is not an issue.
I had just watched this video, and later that same day my A1200 power supply (aftermarked) died on me. So good timing! The Bahar case company does not ship to north America. Does anybody know of a good enclosure available?
Hmm, thats a strange IEC inlet with fuses in both Line and Neutral. It is generally redundant to fuse the Neutral, the only time you would really require this would be if your 230V supply was actually split phase i.e. both line and "neutral" are 115V 180 degrees out of phase (as it may be in the US for example). This generally doesn't happen in 230V countries, and Neutral is at "ground" potential. Still, it doesnt hurt :)
@@talideon Good point, I forgot about this. Here in the UK our plugs are polarised, so unless someone messes up the wiring, Line is always 240V and Neutral is always 0V (with respect to ground)
@@jaycee1980 I hear most installations there are nowadays done by shoddy kitchen constructors after a 2-day online course in electrics so I would not trust the polarity being correct too much at all either.
@@benbaselet2026 well i dont use my Amiga in the kitchen ;) but good point.. though i would prefer those people be BANNED from doing any kind of electrical work
Again really nice work! But why don't you build one universal power supply having +/-12V, 5V, 9VAC (and maybe 3.3V?) and connectors with all these voltages at the front or back of the housing? Then you just need to build a matching cable for the different systems and the power supply could remail on the shelf in front of you.
I'm actually working out a plan for a truly universal one at the moment. So far, I've been using my machines in different spots, so the separate PSUs were the better way of doing it for my use case.
How about a little black paint on the silver ground screws? Nice job. You should see how I hacked an ATX power supply into my Amiga 4000, long ago before I knew about the Meanwell.
Hi Jan, thanks for the video, great as usual... I've seen many people fitting those Meanwell power supplies inside the original enclosure instead of trying to fix the original circuit so I have a question for you - is that really so hard to fix the original Amiga Power Supply circuit? I'd love to find a tutorial on how to properly do that and keep the original psu circuit board.
It's not hard, but generally there is no point. Most of the power supply designs are minimal cost, not efficient, and not reliable by modern standards. There's also very little protection for things like overvoltage in them. Even if you only recap them, by the time you have spent the money on good quality capacitors, you have paid for a Meanwell supply which is superior in every way.
There are several different designs of these. Some are repairable if you really wish to use them. I'd suggest adding a crowbar protection on the output to protect your machine in an overvoltage situation.
Thanks! I only realized that I mixed up ale (which isn't necessarily a bad thing) and awl again when I edited the video. Happens to me all the time. :D
You seem to like to use single-pole switches, I have seen you use those in previous builds. Would be best to switch to two-pole ones (4 terminals) that break both wires at the same time, since our wall plugs (we use Schuko plugs in spain) are not polarized so there's no assurance that L and N are actually L and N.
Technically true, but hardly meaningful in this case. No circuit no power and there's not too much that could go wrong inside. If someone is dumb enough to go poking inside when the plug is inserted then that's a whole new level of problems.
@@benbaselet2026 Well, yes but no. If the power supply happened to short to earth, it could still keep Live shorted even if the switch is off. There's also capacitive effects that could happen, but in this case they are not meaningful.
@@gigaherz_ Sure, but not many wires to do that and he does have a fuse there with solid grounding all over so a rather acceptable risk with reasonably good wiring.
Nice ! Very neat. My version uses a mini-itx power converter in a box and an external generic "laptop style" 12V supply, seems to work very well. I actually have no idea why people persist with the original supplies or pay so much for ready made "new" ones - pretty simple to make as you have showed, all you need is the specific plug for the amiga end and some really low cost parts :)
Cost more to aquire the parts and build this than it does to buy a "new" one on ebay. And that's assuming you have all the necessary tools already in your posession, if not it'd be way more expensive
Let´s face it: the "shielded" Meanwell Rt 50B/RT 65B is a little too big to fit in an original Amiga Powersupply-case. An RPT 65B is unshielded but quite more compact. This model delivers the same voltages (with a little less amps on both the 5V and 12V lines), but it is still much more you will ever need for your Amiga. The main advantage over the Rt 50B/RT 65B is, that the RPT 65B fits into that kind of case without issues. If you have a 3D-printer, there are even some stl-files for an adapter plate available from thingiverse to fit them easily into an original Amiga powersupply-case...Since I have my "Tannenbaum"-drill I only fit round-shaped switches or plugs into my cases :)
One thing I wish the A500 had was an internal PSU like the Atari STf/e and Falcon030 had. But then, with how shady some of Commodore's electronics were, an Internal PSU might have been inviting disaster.
I would like you to measure and compare ripple from Meanwell and original Amiga psu. Jeas from Icomp constantly warns about it, and fact is that some ACA accelerators works unstable with Meanwell. It would be nice to know this for a fact finally.
@@anakondase I experienced differemces on overclocked ACA. My friend as well. Now, I repeat, ir would be great if someone on YT scientifically compare / demistify this.
Nice build, but a few notes... At the end of the day if one needs to buy a case in addition to a Mean Well supply, it's probably less expensive and more convenient to buy a completed aftermarket finished supply that uses a similar Mean Well, like the ones from Electroware. If you already have a case at no cost such as re-bricking the A500 PSU case, then the Mean Well RPT-60B has great specs and is much smaller than the RT-65B, intended for putting inside a plastic shell.
A tip for directly drafting onto the material, put some painters tape down and draw on that. If it's not what you want then just take the tape off and put some fresh tape down. Once you have what you like, you can also use a knife (or your calipers) to cut the tape where the hole goes. When cutting and filing the holes in the material, leave the tape on and use the tape holes to get the cut edges straight. Cheers,
Real-life ripple is around 30mV p-p on the 5V rail. I understand there are some older versions of the 65B where the ripple was a huge problem. This newer models seems to be absolutely fine in that regard.
Do not know whether you want to call it criticism by any kind but I would have tried to use the original case just for retro' s sake. I see that your case is pretty damaged after your efforts with the rt 65b, which, in my opinion, is not fitting really well into the original Amiga case , so I understand why you had to put it into a larger, brand new case. There is a smaller , alternative model , I think it it the rpt 65b from meanwell available which fits much better into an original Amiga powersupply case which, in hindsight, would not have lead to those issues, well, in hindsight...
Yeah, hindsight. I didn’t know at the time that it wouldn’t fit and messed it up. Trust me, I’m most unhappy with that. I‘m still holding on to the original case and am probably going to put it to good use at some point (fixing the holes and painting it is a plan).
No idea why they did that in the original PSU cable. The blue wire is hand soldered to the shielding in the cable from factory. The ground wiring I added on the mains side is all yellow/green which is the current standard here in Germany.
I'm getting REALLY ANNOYED with the misinformation Jens is spreading on Twitter in response to this video. So here's a few points:
Jens claims Meanwell's are bad, but original Commodore PSU's are good. What he doesn't mention is that for most of the PSU's Commodore shipped - They are EXACTLY THE SAME TOPOLOGY. The only difference is the Meanwell's use more modern controller IC's, and probably a higher switching frequency. They also implement undervoltage lockout, and overvoltage shutoff - which none of the original C= ones did.
He claims that the Meanwell's do not compensate for voltage drop in the cable. That's correct... but NEITHER do the original C= PSU's. If they did, they would have an extra wire on the +5V line for remote sense. One or two of them even have a preset pot inside for adjusting the +5V output - exactly the same as the RT-50B and RT-65B do.
Jens's design doesn't have remote sense either. He is "faking it" by measuring the current draw on the 5V line *at the PSU* and then boosting the voltage when more current is drawn. That works fine assuming he has characterised the voltage drop of the cable on the PSU, but here's the thing - the cable drops negligible voltage. Most of the loss is happening in the pins of the PSU connector. The only thing that helps with this, is contact cleaner.
He claims that the regulation feedback from the Meanwell's is not accurate because it's not on the +5V line - IT IS on the +5V line. He is thinking of ATX PSU's which quite often only take feedback from the +3.3V and/or +12V rails. The Meanwell's also use the same TL431 voltage reference IC that countless PSU's, including all of the "off line" switchmode supplies that Commodore shipped.
He's claimed that Jan's PSU connecting the shield ground to mains ground is incorrect. Guess what - that's exactly what the original Commodore PSU's do as well... want to confirm that ? Get a multimeter in continuity mode, put one probe on the outer shield of the Amiga plug, and another on the Earth pin of the mains plug. It will beep and show practically 0 ohms reading.
Finally, he says the ripple and noise of the Meanwell's is not good enough for Amiga use. This is demonstrably untrue... recently "Retro Theory" has done just this on his Twitter account. They meet or exceed the ripple and noise produced by the original Commodore supplies. He claims people who have done this test have only done so off-load - Retro Theory has clearly done it on-load with an Amiga attached, and I have done it by loading the power supply with high wattage resistors. Retro Theory's test is better as it shows the response with a dynamic load.
(I say "Commodore" supplies - but Commodore had nothing to do with them, they are all subcontracted. All Commodore specified was required voltages and minimum required currents. How the subcontractors did that was up to them. There were several vendors - off the top of my head: Speedy-Tech, ISMET, DCD, DVE, Phi-Hong. None of the designs were anything special, and a few of them were dreadful!)
Never apologize for doing the work you love.. great build Jan
Sitting here at 4am, a stomach bug got me. Sipping a hot chocolate and watching Jan. I feel better now lol
Great video Jan!!!
Dude your drilling technique is scary😂
A past sin revisited and rectified. :) That new case does look so much better, and I see no problem with a mostly new unit not masquerading as an original part. If I ever decide to take the plunge and build a PSU, I'm definitely revisiting your videos first. Thank you for sharing your knowledge and experience!
very nice build! May I suggest that you check that earth connection actually made it to the metal frames themselves? Unless I'm mistaken you checked continuity to the screws but there is a small chance that the metal (painted) cases are actually not making contact with them. Unlikely but better safe than sorry! :) You'll have to scrape the paint on a hidden corner to check. Thanks again for the nice video!
good point, added a comment, 🌩⚡🔥🙂
Yes, ideally you should sand away the powder coating and/or use a star washer to "bite" into the metal for ground points
@@jaycee1980 exactly. I got a Chinese pre-heater board from Ebay and it came unearthed. The wires were there but were attached to painted surfaces. I sanded the coating down and, as you say, added a star washer to bite into the metal.
I was thinking same thing. I always sand the paint off where the earth lug is going to be connected.
Me: "Jan should just use a piece of paper to work out the screw spacing."
Jan: "I have this piece of paper I found"
Me: "YAY!"
The backside of the plastic plates have grid pattern on them, quite useful if you want square holes that aren't at an angle.
I keep forgetting that and only ever realize it after I’ve cut the plates. :D
Great video. I've been deep in Commodore power supplies for the past few months so your timing is impeccable! 🤓
The icomp psu was my only option to run my a1200 ppc with bvision, SubwayUSB ect stable. Original C= psu began to shut off my system, atx psu damaged the Subway Controller, the system refused to start with another meanwell-based psu (but worked with my unexpanded a500) A neverending story.
For me the icomp psu solved my problems.
For this built: looks very nice👌
Great work!
Yes, the iComp PSU is a magnificent piece of work that combines 100% compliance with original Commodore specs and still delivers more power. I can imagine that it is the go-to solution for some more elaborate setups indeed. The Meanwell is a simple solution that works for most generic setups (and my slightly expanded Amigas). I have high respect for Jens' work on the iComp PSU, I actually chatted about it with him a while back and he explained the specs he had to work with to me. Excellent work by him and his team indeed even if it seems a bit over engineered (but for good reasons apparently). :)
Excellent job Jan. Looks fantastic and performs as well!
If only those Bahar enclosures were available here in the USA. However, I did make my own Amiga power supply with a plastic enclosure and the same Mean Well RT-65B. I didn't have an original PSU to start with, so I used a square DIN connector that I bought off of Ebay, some vehicle trailer light wiring for the DC side, a normal toggle switch from the local auto parts store, and a power cord from a dead power strip for the AC side. It worked perfectly, thanks to Jan for his videos!
This looks great now! I also have a project box, having the same switch and a very similar Meanwell power supply to yours and it's working very well for my own Amiga.
Great job Jan! You are the "emperor" of power supplies ;-)
It looks very nice, very professional. One of your best yet I think. 😊👌
Great follow-on to the prior video! I bought an inexpensive manual (not electric) nibbler a while back and haven't used a Dremel since for this kind of work :)
A nibbler has been on my shopping list for ages. I need to get one sometime, I guess. :D
22:00 ... but... but... there are some of the connectors you are looking for, in the top left compartment of that box...! or are those the wrong size?
They are too big unfortunately. I used up all the ones in the correct size for my previous power supply builds... :D
Jan,ho sempre detto che sei un genio,sei una persona molto precisa,e l'alimentatore AMIGA ne e' la conferma,il lavoro e' eccellente
What a power beauty with a boing on top! Nice job.
You make me nervous dremelling without safety glasses. Part of the fun of your videos. ;)
I got too excited about the build and forgot. I literally had my safety glasses sitting at arm's length from me during dremelling... :D Sorry.
Very neat build. I liked that you made the ground as a continous chain.
Very neat PSU build, you're definitely getting better at these ;)
No blood, no swearing, it's a win
I'm using the same PSU for my A1200. I had a longer lead to the Amiga and practically no load on the 12V which floated the 12V too high since I had to adjust 5V a bit higher to compensate for losses so that I still have 5,0 volts on the computer board. Fixed it by jumping the +12V out via a beefy diode to the NC terminal, added a load resistor from NC to COM and took the +12V out from the lowered voltage NC terminal. I expect the 12V goes up more and more as you load the 5V with more amps and the switcher does more work.
Oh, good point. I’m going to keep an eye on that. According to the datasheet, none of the rails needs a minimum load but I guess that’s not the whole truth in this case. Good to know.
Hey Jan, will you put the old box on ebay and have it signed by you?
well done Jan Beta really great video very entertaining i love watching your channel
I built a new computer a year ago. I needed a dremel and a drill for it. I bought dual 360mm radiators. The top one didn't fit because the hole was too small. So I dremeled it out. But then the water pump had no place to attach so I drilled into the hard drive cage. It turned out really well. At the same time I drilled into my C64 so I could finally have a proper mount for my VIC-II² switch. Also turned out well. Surprisingly the plastic on my C64 is in amazing condition. It's very soft not brittle. The dremeling of the case though, that took 2 cutting bits. It's a thick steel case. If I needed to do anymore I would have used a 3rd as the 2nd one was pretty dead. lol Also I didn't have a stepper drill bit so I just started small and kept going bigger and bigger until everything fit. lol
I think my A1200 PSU is on its way out, it keeps freezing until it warms up for a few minutes. So, I think I'll be doing this build soon. Not sure to re-use the PSU case or buy/print one. Oh and a couple of step drills are among my favourite tools - I can't believe I didn't get them sooner. Great work Jan.
Nice looking power supply. Much better than the original one. Well done Jan.
Your build skills have for sure improved over the years, I'd say your earlier builds look like "Bastelwaren", this one definately has a nice "German" look and looks like it came out of a factory.
7:28 2A would be the max input @100V (Japan) - fusing a 65W supply with 3,15A for 230V (=725VA) seems a little off, although the data sheet mentions an inrush current of 50A at cold start.
The internal mains fuse in the Meanwell PSU is a 4AT polyfuse, I guess I’m not off by that much? I’m not an expert though, obviously. The extra fuses on the IEC socket are just kind of an extra layer of protection in case some of my wiring shorts out catastrophically. :D
The inrush current is what flows when you turn it on "cold".. but the current flow lasts for a couple of milliseconds at most. This is why a T-type (time delay) fuse is used, it will hold for some seconds before blowing. In the case of a switch mode power supply, the inrush current is caused by the large 400V capacitor charging from empty. Certain kids of transformer, notably toroidal transformers, have an inrush current too.
@@jaycee1980 So the cap is there to make the inrush a non-issue and to provide lower noise and more constant current down the road ? I'm not an engineer so bare with me. :)
@@VincentGroenewold No, the capacitor is what causes the inrush. In a switch mode power supply, the mains voltage gets rectified and smoothed into a high voltage DC supply which feeds the primary side of the transformer. When this capacitor is empty, it draws a large gulp of current on power up as it charges.
edit: to make this clearer, since as people will say "But transformers dont run on DC" - what happens is the mains voltage (which is 50-60Hz AC) is smoothed into ~380V DC, which is then "chopped" by a MOSFET into high frequency AC (usually anything from 50-500KHz), which drives the transformer. The reason is that the higher frequency means the transformer core can be much smaller than the equivalent core needed for 50-60Hz. This is why SMPS supplies are smaller and more efficient.
@@VincentGroenewold The capacitor is the main thing that cause the inrush current when started "cold" (it has to be charged), but it's necessary for proper function. The input capacitor needs to scale with the max output power (it needs to be able to provide power for a set amount of time) and if it's this is not managed in some way the inrush current scales linearly with the capacitor size.
If you look at bigger PSUs they often have something to reduce the inrush current, like a NTC resistor or an inductor, and often on higher-end models a relay is used to bypass it once the in-rush has pased to avoid the efficiency loss (it's only desirable for that split second during cold start). The inrush current is also often very different on 110V and 230V - it's tends to be a much bigger problem with 230V since naive/simple constructions (like this PSU) often end up with 3-4 times larger inrush current on 230V compared to 110V, the RT-65 series doesn't even bother saying what the inrush current is on 110V.
The RT-65 series says that typical inrush current at 230V is 50A which is a significant inrush current for a PSU this small, as a comparison I tend to want 600-850W PSUs to not go above a similar 50A-60A inrush current! and those have WAY bigger input capacitors (~10x) so they'd probably draw 400A+ without "in-rush protection" which would likely immediately trip your breaker even with nothing else connected! IIRC 60A or 70A inrush current is where Tom's Hardware start mentioning "high inrush current on 230V" on the downsides (IE warnings) section for a PSU test.
Managing inrush current can be important when you have many *or* powerful PSUs, if your breaker immediately trips when trying to get power back either something broke or you have an inrush current issue. If you can get all of it working using the "unplugging dance", IE unplug you "enough" that you can reset the breaker and then plug stuff back in one at a time to spread out the "spikes". If that worked the the issue is that the combined inrush current is too large.
I suspect there's no inrush protection at all in this unit which they only get away with because it's "only" 65W (so it has small capacitors) and is cheap. The 80mV ripple on 5V isn't fantastic either but it's the max possible value (so likely better or much better) and is likely fine for anything other than high-end audio anyway! Certainly I wouldn't expect any issues with the hardware he's planning to power with it.
The 50A inrush current per RT-65 does mean it's probably not the best PSU to use if you want to say use a bunch to power a whole fleet of machine for an exhibit for example, but it's unlikely to be an issue in most other scenarios.
very nice work, very nice power supply, and once again a very nice video !
Nice, though I would probably have put some rubber (or plastic) pad/feet between the PSU and the case for vibration dampening and avoiding that "tin" sound if you tap the bottom.
Not really important, just a very minor Quality of Life tweak you can do at any time.
Ooooooooh! Precisely what I want to do as well. Thanks Jan!
Love your work Jan! ☺
Great project. Huge improvement.👍🏻 Greetings from Hamburg✌🏼
Cool looking PSU. :)
Finally this PSU project came out really nice! I also like this case, looks very decent and solid.👌
Maybe I'm going to build a double PSU like this for my Atari 520 STM plus SF 314 Floppy, we'll see. 😉
Looks like no stranded wire for the mains switch? If so that’s definitely no recommendation from me. Unfortunately it’s hard to tell since assembly is always fast forward.
All stranded wire (salvaged from a mains cable).
@@JanBeta Good decision, like I said, it’s hard to tell in fast forward. My Vids are usually very boring since soldering takes forever and is not entertaining but on the other hand I’m too lazy for editing…
Step drills rock.
What size enclosure are you using ?
Added the size to the video description now. I used the 150x70x170 mm version.
@@JanBeta thanks for the quick reply. Great work 👌
And as so often - black just looks good :)
what size enclosure is it?
Added the size to the video description now. I used the 150x70x170 mm version.
Had for a long time a ATX PSU with a nosy fan. But now for 2 years a iComp CA-PSU. Have four of them for my A500 and my three A1200’s. Have to many money in my A500 and A1200’s. Do the safe way and not the dumb way such as my old ATX PSU.
If it's wrong to love step drills, I don't want to be right.
I fitted a similar Meanwell PSU inside my Commodore 1541C drive last year, the linear supply was just insanely heavy and ran hot when idling (mostly trying to get it to work with VICE with a driver that windows refused to install), with the new one, it's light and runs cool, still have the original PSU stashed away though, just in case I ever needed to refit it... :D
Beautiful power supply Jan Beta. I would have gone black m3 nuts and bolts for the earth strap. Just to match the case.
I used the bare metal ones because the black ones are usually painted and don't make contact as well. But it probably wouldn't matter in this case (because of the washers etc) and would indeed make for a speaker look. :)
Should use some "star" washers underneath the screws securing the power supply. The washers reduce the screws' loosening from vibration and cut into the paint for an improved connection to the metal.
Good point! I should add them to those screws, too. :)
Shame Bahar dont ship to the uk from the eBay site
Great video, Jan! I'm thinking of building a universal psu for my retro computer stuff. Basically one psu with all the appropriate voltages and a couple of generic utlets, and then different power cables for the computers. :)
How on earth can the Meanwell destroy the hardware in the Amiga? Please elaborate.
@@JanBeta Jens Schoenfeld says this, but it's because he wants you to buy his product. The Meanwell power supplies are better in every way than the original power supplies which shipped with the Amiga in the first place.
@@djbase_ Which are…. ? I wouldn’t think Meanwell’s are bad for it, I imagine they are even better controlled and lower noise then the original PSU’s.
@@djbase_ As an electronics engineer, I can tell you right now that you are wrong. There is nothing wrong with the output of the Meanwell power supplies that would cause any issue with an Amiga.
@@VincentGroenewold He is repeating the rubbish spouted by Jens. Jens claims that the ripple and noise of the Meanwell's does not meet Commodores spec, and points to an internal memo released by Commodore to prove it. However, in fact the Meanwells have superior specs to the PSU's that Commodore *actually* shipped... so it is not an issue.
Good job.
18:13 Help me step drill. I'm stuck!
I had just watched this video, and later that same day my A1200 power supply (aftermarked) died on me. So good timing! The Bahar case company does not ship to north America. Does anybody know of a good enclosure available?
Hmm, thats a strange IEC inlet with fuses in both Line and Neutral. It is generally redundant to fuse the Neutral, the only time you would really require this would be if your 230V supply was actually split phase i.e. both line and "neutral" are 115V 180 degrees out of phase (as it may be in the US for example). This generally doesn't happen in 230V countries, and Neutral is at "ground" potential.
Still, it doesnt hurt :)
Schuko, while a safe design, is unpolarised, so having two fuses is a good idea. Either non-ground pin can end up being live or neutral.
And it built to use 230v from split eg usa compatible
We use 220-250v here fyi though ofc not with something this low wattage
@@talideon Good point, I forgot about this. Here in the UK our plugs are polarised, so unless someone messes up the wiring, Line is always 240V and Neutral is always 0V (with respect to ground)
@@jaycee1980 I hear most installations there are nowadays done by shoddy kitchen constructors after a 2-day online course in electrics so I would not trust the polarity being correct too much at all either.
@@benbaselet2026 well i dont use my Amiga in the kitchen ;) but good point.. though i would prefer those people be BANNED from doing any kind of electrical work
Again really nice work! But why don't you build one universal power supply having +/-12V, 5V, 9VAC (and maybe 3.3V?) and connectors with all these voltages at the front or back of the housing? Then you just need to build a matching cable for the different systems and the power supply could remail on the shelf in front of you.
I'm actually working out a plan for a truly universal one at the moment. So far, I've been using my machines in different spots, so the separate PSUs were the better way of doing it for my use case.
ua-cam.com/video/FzPQ4dETpwM/v-deo.html This guy just built a common commodore PSU for himself
How about a little black paint on the silver ground screws? Nice job. You should see how I hacked an ATX power supply into my Amiga 4000, long ago before I knew about the Meanwell.
Love it!
Hi Jan, thanks for the video, great as usual... I've seen many people fitting those Meanwell power supplies inside the original enclosure instead of trying to fix the original circuit so I have a question for you - is that really so hard to fix the original Amiga Power Supply circuit? I'd love to find a tutorial on how to properly do that and keep the original psu circuit board.
It's not hard, but generally there is no point. Most of the power supply designs are minimal cost, not efficient, and not reliable by modern standards. There's also very little protection for things like overvoltage in them. Even if you only recap them, by the time you have spent the money on good quality capacitors, you have paid for a Meanwell supply which is superior in every way.
There are several different designs of these. Some are repairable if you really wish to use them. I'd suggest adding a crowbar protection on the output to protect your machine in an overvoltage situation.
@@benbaselet2026 i've seen only one that has such a crobar, uses an L296 regulator IC and a linear transformer. I often find that SCR blown in them!
Amazing job.
Hi Jan. Great vid. Your 'pokey-bit' is called an Awl (pronounced 'all'). Keep up the retro fun ;-)
Thanks! I only realized that I mixed up ale (which isn't necessarily a bad thing) and awl again when I edited the video. Happens to me all the time. :D
Buen trabajo, saludos, Tony
You seem to like to use single-pole switches, I have seen you use those in previous builds. Would be best to switch to two-pole ones (4 terminals) that break both wires at the same time, since our wall plugs (we use Schuko plugs in spain) are not polarized so there's no assurance that L and N are actually L and N.
Technically true, but hardly meaningful in this case. No circuit no power and there's not too much that could go wrong inside. If someone is dumb enough to go poking inside when the plug is inserted then that's a whole new level of problems.
@@benbaselet2026 Well, yes but no. If the power supply happened to short to earth, it could still keep Live shorted even if the switch is off.
There's also capacitive effects that could happen, but in this case they are not meaningful.
@@gigaherz_ Sure, but not many wires to do that and he does have a fuse there with solid grounding all over so a rather acceptable risk with reasonably good wiring.
@@benbaselet2026 Agreed. It's reasonable. I was just pointing out a switch that disconnects both would be ideal there.
Nice ! Very neat. My version uses a mini-itx power converter in a box and an external generic "laptop style" 12V supply, seems to work very well. I actually have no idea why people persist with the original supplies or pay so much for ready made "new" ones - pretty simple to make as you have showed, all you need is the specific plug for the amiga end and some really low cost parts :)
I did the same ;-)
Cost more to aquire the parts and build this than it does to buy a "new" one on ebay. And that's assuming you have all the necessary tools already in your posession, if not it'd be way more expensive
cracking build...........maybe a video of cleaning that joystick 🤣🤣🤣
Yeah, the joystick hasn't been cleaned since I got it in the early nineties... Probably a good idea... :D
Let´s face it: the "shielded" Meanwell Rt 50B/RT 65B is a little too big to fit in an original Amiga Powersupply-case.
An RPT 65B is unshielded but quite more compact. This model delivers the same voltages (with a little less amps on both the 5V and 12V lines), but it is still much more you will ever need for your Amiga. The main advantage over the Rt 50B/RT 65B is, that the RPT 65B fits into that kind of case without issues. If you have a 3D-printer, there are even some stl-files for an adapter plate available from thingiverse to fit them easily into an original Amiga powersupply-case...Since I have my "Tannenbaum"-drill I only fit round-shaped switches or plugs into my cases :)
Hey did one too Jan.
One thing I wish the A500 had was an internal PSU like the Atari STf/e and Falcon030 had. But then, with how shady some of Commodore's electronics were, an Internal PSU might have been inviting disaster.
Atari's frequently have power supply problems too. External is better because you can simply replace it.
Very nice.
I would like you to measure and compare ripple from Meanwell and original Amiga psu.
Jeas from Icomp constantly warns about it, and fact is that some ACA accelerators works unstable with Meanwell. It would be nice to know this for a fact finally.
It isn't a fact. Others have done it. Jens simply wants to sell his own product.
@@jaycee1980I know that we talked about it on FB, but that would be great theme for one scientific video, as Jan knows how to make.
@@lokz9632 I've been using the exact same power supply for my accelerated A500 with no issues att all. I lean towards the same conclusion as Peter.
@@anakondase I experienced differemces on overclocked ACA. My friend as well. Now, I repeat, ir would be great if someone on YT scientifically compare / demistify this.
@@lokz9632 Sounds more like the ACA is the problem to be honest. Badly designed?
So the first power supply was still true to the "Jan Beta Was Here" tradition :) I'm sad we don't get more of these anymore :)
Don't you worry, I'm still going to accidentally destroy things occasionally... :D
Nice build, but a few notes... At the end of the day if one needs to buy a case in addition to a Mean Well supply, it's probably less expensive and more convenient to buy a completed aftermarket finished supply that uses a similar Mean Well, like the ones from Electroware. If you already have a case at no cost such as re-bricking the A500 PSU case, then the Mean Well RPT-60B has great specs and is much smaller than the RT-65B, intended for putting inside a plastic shell.
A tip for directly drafting onto the material, put some painters tape down and draw on that. If it's not what you want then just take the tape off and put some fresh tape down.
Once you have what you like, you can also use a knife (or your calipers) to cut the tape where the hole goes.
When cutting and filing the holes in the material, leave the tape on and use the tape holes to get the cut edges straight.
Cheers,
Winner Winner Chicken Dinner
that sticker reminds me of mad balls
Wow...
He said he was going to edit out all the swear words, so how much longer would the video be with all the swear words left in?
The combined footage I shot for this video was around 2 hours. But not all of the parts I left out were swearing. ;)
You could build a little cottage industry for yourself with these power supplies Jan! :)
I'd love to do that but I can't afford the necessary permits to legally sell these here. :D
Dood somehow this kind of reminds me of going to the dentist.
So you made it worse by ditching the original case? 😁
oh my god jan..... there is the top left box is full of female spade connectors !!!!!!!!!!
Wrong size.
Not too old...
look at the ripple. :(
Real-life ripple is around 30mV p-p on the 5V rail. I understand there are some older versions of the 65B where the ripple was a huge problem. This newer models seems to be absolutely fine in that regard.
@@JanBeta sure does.
BETA Industries strikes again 🌝
Please elaborate. Any specific criticism or just trolling?
Do not know whether you want to call it criticism by any kind but I would have tried to use the original case just for retro' s sake. I see that your case is pretty damaged after your efforts with the rt 65b, which, in my opinion, is not fitting really well into the original Amiga case , so I understand why you had to put it into a larger, brand new case. There is a smaller , alternative model , I think it it the rpt 65b from meanwell available which fits much better into an original Amiga powersupply case which, in hindsight, would not have lead to those issues, well, in hindsight...
Yeah, hindsight. I didn’t know at the time that it wouldn’t fit and messed it up. Trust me, I’m most unhappy with that. I‘m still holding on to the original case and am probably going to put it to good use at some point (fixing the holes and painting it is a plan).
Looking at you use pens and other Drawing implements triggers me... Please learn how to hold a pen properly. It will save your hands.
Been doing it that way for 40 years, would be difficult to relearn. My hands are fine, too, don’t worry. 😉
WHY have you got BLUE as ground??????? it should be green / yellow.... blue is neutral in the UK.....
No idea why they did that in the original PSU cable. The blue wire is hand soldered to the shielding in the cable from factory. The ground wiring I added on the mains side is all yellow/green which is the current standard here in Germany.
First 😂🍻