Updated Beeb PSU

discuss both original and modern hardware for the bbc micro/electron
User avatar
MartinB
Posts: 5285
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: Updated Beeb PSU

Post by MartinB » Sun Jun 05, 2016 5:50 pm

No worries Ian - if this Mean Well PT-65A turns out to be suitable, I shall in due course return all 157 of said Astecs duly repaired :D

(with an invoice for a touch over £3100 plus postage... :shock: :wink: )

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sun Jun 05, 2016 8:53 pm

Should have gone for 160, he gives a discount at this quantity :lol:

User avatar
sweh
Posts: 2166
Joined: Sat Mar 10, 2012 12:05 pm
Location: New York, New York
Contact:

Re: Updated Beeb PSU

Post by sweh » Mon Jun 06, 2016 1:58 am

I dunno; 20 quid per PSU ain't a bad price!
Rgds
Stephen

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sat Jun 18, 2016 11:30 pm

So having mentioned power meters earlier in the thread, I finally got both of mine out of storage. So here are the results...

Acorn BBC B (issue 7 PCB) with a modern sideways ROM/RAM board (IFEL) and disk drive interface chips fitted

Prodigit model 2000MU power meter:
Supply voltage 244.4V at 49.9Hz, current 0.13A, power 20W, 31VA, PF=0.64

brennenstuhl PM230 power meter:
Supply voltage 241V at 50Hz, current 0.12A, power 28.92W

Acorn BBC Master 128 with no expansions, but twin 3.5mm drives connected (but not being accessed)

Prodigit model 2000MU power meter:
Supply voltage 244.4V at 49.9Hz, current 0.08A, power 14W, 20VA, PF=0.66

brennenstuhl PM230 power meter:
Supply voltage 241V at 50Hz, current 0.09A, power 21.6W

Mark

User avatar
MartinB
Posts: 5285
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: Updated Beeb PSU

Post by MartinB » Sat Jun 18, 2016 11:56 pm

I'm a bit confused assuming I've read your post correctly.... :?

There seems to be a big difference in the results between the two meters? For the Beeb, one says 20W with the other showing nearly 29W and for the Master, it's 14W versus 22W ?

For the Beeb +5v consumption (DC output power, not AC input power as you are measuring), I arrived at just under 11W and assuming your IFEL board doesn't add much and neither of us are using +12v, your measurements combined with mine therefore imply that the Acorn SMPS has an efficiency of around either 55% or much worse at 40% depending on which of your two power meters is correct.... :-k

poink
Posts: 963
Joined: Tue Mar 01, 2011 10:27 am
Contact:

Re: Updated Beeb PSU

Post by poink » Sun Jun 19, 2016 3:51 am

MartinB wrote:There seems to be a big difference in the results between the two meters? For the Beeb, one says 20W with the other showing nearly 29W and for the Master, it's 14W versus 22W ?
They're pretty much the same figures - it looks like the 14W one is correcting for a power factor of 0.66 (lovely, lovely, switch mode harmonics!), and the 22W one isn't. (22W * 0.66 = 14.52W)

That means that 22W is the true power draw, however, AIUI 14.52W is what you're paying for at the meter - from memory, in domestic settings, it's generally assumed that power factor is 1).

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sun Jun 19, 2016 7:42 am

Yeah, for low power loads, these power meters are not that accurate. They are really designed for working out the energy use of heaters, cookers, irons, washing machines and such like. Not 1980's computers, or gadgets and gizmo’s :D

For the testing, one power meter was plugged into an extension socket, then the second power meter was plugged into the first one, then finally the computer was plugged into the second power meter.

The true consumed power is the lower figure, as that is the energy used by the load. This is equivalent to the power dissipated in a purely resistive load like an electric fire or a tungsten filament lamp. The higher "reactive" power figure is just the measured voltage times the measured current. But, during the cycle, some power is returned to the mains, hence the true power consumed being the lower figure.

This is the reason why you need to know what and how your are measuring an electrical quantity, rather than just blindly reading a displayed value and then believing the meter to be telling you the "true correct" figure :lol:

Electricity companies normally only charge for true power consumption for domestic customers and small offices. If the meter does record power factor, they do not use this information. Only some modern digital domestic meters record power factor. Nearly all of the older electromechanical meters only record true power.

Electricity companies prefer the power factor to be close to 1. As lower power factors (higher reactive current levels) require more power station capacity, thicker distribution cables, larger sub-station transformers etc. etc.

Mark

User avatar
MartinB
Posts: 5285
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: Updated Beeb PSU

Post by MartinB » Sun Jun 19, 2016 7:48 am

Ah ok, thanks - I saw you had put a 0.6x factor for the first meter but because you didn't put anything for the second, I assumed you'd be quoting like with like and I didn't think to try the conversion.

Anyway, without actually knowing the PSU efficiency up front, we can't say for sure that your input figures align with my output current measurements but as I said earlier, if we do assume the machines are similar enough, then the Acorn PSU has a fairly low efficiency compared to say the Mean Well PSU that tricky spotted which claims 76%. Is that a fair assumption?

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sun Jun 19, 2016 7:54 am

I just quoted the displayed values and then waited to see how long it would be before someone commented on how different the power results were :lol:

As poink says, the brennenstuhl meter is displaying reactive power, so it should be "VA" not Watts.

Mark

User avatar
MartinB
Posts: 5285
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: Updated Beeb PSU

Post by MartinB » Sun Jun 19, 2016 8:04 am

Ha ha, I'm not interested in playing science games ta, I was trying to put some numbers to my findings that the 'new' composite PSU I built appears to run a lot cooler than the original Acorn PSU despite the enclosed cases. A lower efficiency means more heat for the same output power and it looks as if your numbers (ignoring the trap :wink:) do confirm this... 8)

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sun Jun 19, 2016 8:09 am

MartinB wrote:Anyway, without actually knowing the PSU efficiency up front, we can't say for sure that your input figures align with my output current measurements but as I said earlier, if we do assume the machines are similar enough, then the Acorn PSU has a fairly low efficiency compared to say the Mean Well PSU that tricky spotted which claims 76%. Is that a fair assumption?
I would have expected the Beeb power supply to be around the 60% efficiency figure myself. If the Beeb power supply was a lot better than this, it would run cooler. Also keep in mind, when you see a figure for efficiency, it's often the best efficiency figure at a specific load.

And, as I said in my earlier post, these power meters are not very accurate when measuring low power loads. So they could be over or under reading the true power value.

Mark

User avatar
MartinB
Posts: 5285
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: Updated Beeb PSU

Post by MartinB » Sun Jun 19, 2016 8:14 am

I noted 55% for the Acorn PSU using your true figures with my measured ones so that's probably about right then. I would expect modern SMPS to be better so 76% sounds easily achievable at Beeb loads versus the PSU capacity and I suspect that the APD I'm using isn't far away from that either.

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sun Jun 19, 2016 8:23 am

If someone would like to donate some suitable Beeb PSU output cable -> motherboard power connectors, so that I can make up a test cable (or a surplus cable loom from a Beeb power supply), I can carry out a full range of voltage and current measurements.

I don't think I have any of that type, having never needed to replace any so far :mrgreen:
And I'm not sure if I can jerry rig a test lead for three 5V connections, plus a -5V connection :? :-k

I can borrow a nice expensive (all singing and dancing) Fluke multimeter from work that has true RMS voltage and current ranges, to measure the mains input as well.

Mark

User avatar
MartinB
Posts: 5285
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: Updated Beeb PSU

Post by MartinB » Sun Jun 19, 2016 8:32 am

tbh Mark, I don't really think we need any more detail than we now have and I'm not sure what practical value it would add? I'd say what you've done already is sufficient combined with my actual DC current measurements. Everything seems to stack up and my additional temperature monitoring suggests that we can do better PSU-wise in the Beeb space envelope without any particular worries. I'll probably give one of those Mean Well a try at some point when I have a suitable time window.... :)

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sun Jun 19, 2016 9:06 am

No worries :lol:

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Sun Jun 19, 2016 5:03 pm

These are the two power meters that I used:
image.jpeg
image.jpeg
image.jpeg
Mark

User avatar
sweh
Posts: 2166
Joined: Sat Mar 10, 2012 12:05 pm
Location: New York, New York
Contact:

Re: Updated Beeb PSU

Post by sweh » Sun Jun 19, 2016 5:13 pm

For whatever it's worth, my modified-to-US-voltage model B says 22W, 36VA, 0.3A, 0.61 PF according to a "P3 Kill-A-Watt" ( https://www.amazon.com/P3-P4400-Electri ... B00009MDBU ).
Rgds
Stephen

User avatar
topcat96
Posts: 268
Joined: Thu Jun 26, 2008 1:17 am
Location: Somewhere wonderful!
Contact:

Re: Updated Beeb PSU

Post by topcat96 » Mon Jul 04, 2016 3:47 pm

i know this is more than a plea than wishful thinking, but is anyone here making/testing or designing replacement BBC B power supplies to hopefully go on sale to those who need them?

The reason i'm asking is that I have 3 BBC's that are in dire need of replacements, but at present they are nothing more than mere ornaments on the shelf (nice looking ones though!).
Image

User avatar
daveejhitchins
Posts: 5518
Joined: Wed Jun 13, 2012 6:23 pm
Location: Newton Aycliffe, County Durham
Contact:

Re: Updated Beeb PSU

Post by daveejhitchins » Mon Jul 04, 2016 7:06 pm

topcat96 wrote:I have 3 BBC's that are in dire need of replacements, but at present they are nothing more than mere ornaments on the shelf (nice looking ones though!).
If you still have the PSUs (?) then contact Mark at RetroClinic - He'll fix them for you.

Dave H :D
Parts: UM6502CE, GAL22V10D, GAL16V8D, AS6C62256A, TC514400AZ, WD1772, R6522, TMS27C512, AT28C256
Products: ARA II, ARA III, ABR, ATI, AP6, MGC, AP5 . . .
For a price list, contact me at: Retro Hardware AT dave ej hitchins DOT plus DOT com

User avatar
tricky
Posts: 4253
Joined: Tue Jun 21, 2011 9:25 am
Contact:

Re: Updated Beeb PSU

Post by tricky » Mon Jul 04, 2016 7:54 pm

If you don't have any PSUs and are feeling adventurous, you could try the MeanWell from higher up the thread.

User avatar
topcat96
Posts: 268
Joined: Thu Jun 26, 2008 1:17 am
Location: Somewhere wonderful!
Contact:

Re: Updated Beeb PSU

Post by topcat96 » Mon Jul 04, 2016 8:37 pm

daveejhitchins wrote:If you still have the PSUs (?) then contact Mark at RetroClinic - He'll fix them for you.

Dave H :D
Sadly the PSU's are no longer around. They were thrown out by accident when I had a major clear out of stuff last year
tricky wrote:If you don't have any PSUs and are feeling adventurous, you could try the MeanWell from higher up the thread.
Yeah, was looking at those earlier but my soldering and electronics skills aren't exactly brilliant - I'd most probably end up causing a fire or destroying the BBC's completely

No, best to wait I guess and hope somebody with better skills than me wants to take this project on [-o<
Image

User avatar
MartinB
Posts: 5285
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: Updated Beeb PSU

Post by MartinB » Mon Jul 04, 2016 8:53 pm

Just for info, my thread topic APD-based PSU has happily clocked up hundreds of hours already through some long term I2C playing serious research but, recognising said bricks are not an all-comers solution, I do fully intend to give the Mean Well PT-65A a try unless someone else wants to be my guest and go first? It'll still be two or three weeks though because I'm just on with re-lifeing my posh Master so I can finally finish the eeprom 'Holy Grail' story. (In case anyone hadn't noticed, my AT28C256 implementation never worked in a Master, only in Beebs and Elks - it's probably something simple but I just haven't had a stable machine to work with and now I have! :D )

So, perfectly happy to try out tricky's great PSU spot =D> but there'll still be a little bit of a wait I'm afraid..... 8)

User avatar
tricky
Posts: 4253
Joined: Tue Jun 21, 2011 9:25 am
Contact:

Re: Updated Beeb PSU

Post by tricky » Fri Nov 15, 2019 11:15 am

I was just about to buy a PT65A for my PSU-less master and noticed that there is a Meanwell option that I hadn't seen before:
The RPT-60A, it is the same price as the PT-65A, but much smaller and "Medical Grade"/"Medical safety approved" - whatever that means.
The headlines are: +5V @ 4A -5V @ 500mA +12V @ 2A as opposed to a "B"
MartinB wrote:
Wed Jun 01, 2016 8:30 pm
... the standard PSU totals +5v @ 3.75A, -5v @ 100mA and +12v @ 1.25A...
and 101.6 x 50.8 x 29mm or 4" x 2" x 1.14".
https://www.powersuppliesonline.co.uk/p ... upply.html Data sheet is linked on page. RTP-60 series datasheet.
I don't know if this would be any easier to fit, or as we now know the PT-65A, wether it is worth investigating, but I thought that I would check here :)
The 100% load input voltage (60Hz) is 115V vs 100V and Ambient temperature 40C rather than 45C and as I think 1024MAK said, it will be using more of its capability, so it may age faster - but I don't really know how much difference that will really make.

User avatar
1024MAK
Posts: 9906
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Updated Beeb PSU

Post by 1024MAK » Fri Nov 15, 2019 1:26 pm

The PT-65-A is this one.

Either should be fine for a normal Beeb. If you have an excessive number of expansions that draw power from your machine, then obviously you would be better off with the higher power unit.

Mark

Post Reply

Return to “8-bit acorn hardware”