I would install the fuses after the caps; between caps and amplifier. This way the charge current won't go through the fuses and they're less likely to blow due to charging caps.
As I said, this is the best solution at all for a conventional PSU. Remember the supply caps on the amp's PCB.
Best regards!
Best regards!
In BOM list capacitors C8 & C9 are:-
PANASONIC ELECTRONIC COMPONENTS ECOS2AP472DA Electrolytic Capacitor, 4700 µF, ± 20%, 100V, Dia. 30 mm, Height 50mm, lead spacing 10mm, ESR 0.071 ohm Farnell Code no. 1198728
As these are 'No Longer Manufactured', Farnell offers 'Alternatives Available' - Code 9452974 - Electrolytic Capacitor, Snap-in, 4700 µF, 100 V, LPR Series, ± 20%, Quick Connect, Snap-In, 30 mm - £7.54 each, ESR rating not given.
MCLPR100V478M30X50 - MULTICOMP - Electrolytic Capacitor, Snap-in, 4700 µF, 100 V, LPR Series, ± 20%, Quick Connect, Snap-In, 30 mm | Farnell element14
Last edited:
I have used a couple of bigger 40Vac transformers. 800VA and 1000VA.
Both were 230:0-40, 0-40Vac and both gave +-58.5Vdc on the smoothing when mains was @ ~245Vac
If you were to rectify each secondary with it's own bridge, you would get a slightly lower output voltage and guarantee <60Vdc even @ 253Vac.
each diode has a Vf that varies with the current draw.
at near zero current the Vf could be as low as 0.5V, but for the current pulses that pass through when powering a ClassAB amplifier that is ticking over, expect 0.65V to 0.7V per diode. at very high currents for example when starting up and during high power testing the Vf could exceed 1V per diode.
that gives 1.3V to 1.4V per bridge.
I have built and tested both. I have not found any evidence that one is quieter than the other.
A dual secondary can be converted to centre tapped, whereas centre tapped cannot easily be converted back to dual.
No, the current through each is the same. The dual rectifier doubles the losses and has double the devices to dissipate those losses. So each runs at the same temperature.
Fusing between the transformer and the first stage smoothing is a complete waste of resources and effort.
Fusing after the first stage smoothing can be made to work, But I believe that adopting a close rated mains fuse provides sufficient safety for your transformer in event of some abuse downstream.
A mains fuse for each transformer. and make each close rated.
eg
625VA 230Vac needs a T2.7A fuse so use a T2.5A or T3.1A with a soft start that does not blow that fuse during the first 1000 cold start ups (that's less than 3years use).
12VA auxiliary transformer needs a 0.052A Fuse rating, just use a T500mA and start it direct online.
You could even use an F500mA fuse and it will survive start up as well.
Last edited:
It's looking good for the Elektor Q-Watt kits: they are ordable again with availability in March: https://www.elektor.com/q-watt-110656-71
Good news!
Best to place an order now: -
£81.95 or Elektor member price £73.95
In my personal opinion Elektor need to take on board the interest and demand the Q-Watt has and reissue an updated write-up of the whole project.
Best to place an order now: -
£81.95 or Elektor member price £73.95
In my personal opinion Elektor need to take on board the interest and demand the Q-Watt has and reissue an updated write-up of the whole project.
Last edited:
Update.
Yesterday on another Thread, two members reported that they found the dual sec/dual rectifier to give a quieter output from their amplifiers.
I'm interested to learn by what is meant by 'quieter output from their amplifiers' depending on the type of rectifier diodes?
Both Members are well respected.
But as I said earlier I have not seen evidence of quieter on any of my amplifiers.
And one of those Members sold PCBs for a design that is one of the noisiest I have ever built. If that was his quiet version I dread to think what his noisier version would sound/measure like?
Which thread is it, Andrew? I suspect it might be one that I've also contributed to yet, doesn't it?
Ditto. For my understanding, there can't be very much evidence for this claim. You need to be more careful in arranging the ground connections in a single bridge supply, though.
Best regards!
Thanks for all the great info Andrew, Kay + other diy-ers. Especially on the subject of power supply.
I've been away for a week but now looking forward to drilling my heatsinks and mounting first PCB.
Will be interesting to note if Elektor made any changes in this latest release.
FYI - the Elektor Crescendo amp from 1982 I think did specify dual rectifiers - one per rail. There was no guidance or rationale given. Mind you I don't think the PS was the main concern then at Elektor and certainly not now.
Pops.
I've been away for a week but now looking forward to drilling my heatsinks and mounting first PCB.
Will be interesting to note if Elektor made any changes in this latest release.
FYI - the Elektor Crescendo amp from 1982 I think did specify dual rectifiers - one per rail. There was no guidance or rationale given. Mind you I don't think the PS was the main concern then at Elektor and certainly not now.
Pops.
Attachments
Last edited:
Try to upload photo again - always rotated when I post from my phone.
What are the values of those resistors on the caps? On first look they seem a bit small (in wattage).
In the Q-Watt Elektor article is does state the following: -
"The PCB is specifically designed for use as a monaural amplifier (monoblock). For a stereo amplifier you can simply build two of the boards and mount them in an enclosure together with the power supply. You should preferably use two separate power supplies (one for each channel)."
The Q-Watt Measured Performance also states:-
"(Measured with a power supply consisting of a 500 VA power transformer with two 40 V secondaries (Nuvotem type 0500P1-2-040) and four external 10,000 μF, 100 V buffer capacitors)"
And in the parts list it gives us this: -
"Power Supply (for one amplifier) Power transformer: sec. 2x40V, 500VA (e.g. Nuvotem 0500P1-2-040 for 230 VAC mains) Bridge rectifier: 200V, 35A (e.g. GBPC3502) (Fairchild) Four 10,000μF, 100V electrolytic capacitors (2 in parallel on each supply rail)"
Hey Delange. Am using 0.5W resistors, 13kR. Actual dissipation is about 233 mW each when feeding from 55V.
Haven't figured out final location for bleeders yet.
Pops.