You have no holes under those heatsinks, in that case, you supposed to lift up the heatsink from the PC board 5-10mm. Around your heatsinks, the air cannot circulate how it should. For class A/B or low-class A, it would be fine that PS setup, for heavy biased class A amplifier not the best, imagine if you move it into a (almost) closed amplifier case. Be careful when you increase the PS voltage that will increase the heat also. I think you know that. 🙂 Nice build!
I put my PSU board behind the exit of CPU cooler where a breeze from the 27C (still relatively cool) spent air from coolers then pass by diode heatsinks and keeps them a bit cooler than natural convection. This is possible because fans are under the PCB.
Gaborbela,
I was thinking about drilling holes in the board under the heatsinks but, there is no clear shot. I’d either drill through the ground or power plains. Since the to-220 mounting hole is not centered, I’m thinking about flipping the heatsinks around and I’ll get a gap the bottom.
Also, I’m thinking a 500va, 30vac secondaries would be a better option than
28vac. Especially if I go with a CRCRCRC type psu.
Yes, those are super premium looking potted trafos. Nice caps too - what size are those caps? It’s amazing how much money goes into big electrolytics on an amp like this.
Hi Guys
If I may my 2 cents. Since the BB is going to cost a fair bit to build, may I suggest going for Mundorf M Lytic or Nichicon Super Through. Played with fair share of caps & these are the most musical that I’ve tried. The Super Through is rather special, it lets you hear the layers into an instrument, cons its a little slower then M Lytic & it has a slight forwardness that does not work well for my set up. M Lythic very transparent & bass is deep. Rectifier if you can get pls try
D6SBN20 bridge diodes grainless very smooth sounding. The bridge diodes in combi with the psu caps will dictate SQ tremendously. On the amp module try not to use just 1 type/brand of caps . This you guys just have to experiment.
Cheers & Happy Diy
If I may my 2 cents. Since the BB is going to cost a fair bit to build, may I suggest going for Mundorf M Lytic or Nichicon Super Through. Played with fair share of caps & these are the most musical that I’ve tried. The Super Through is rather special, it lets you hear the layers into an instrument, cons its a little slower then M Lytic & it has a slight forwardness that does not work well for my set up. M Lythic very transparent & bass is deep. Rectifier if you can get pls try
D6SBN20 bridge diodes grainless very smooth sounding. The bridge diodes in combi with the psu caps will dictate SQ tremendously. On the amp module try not to use just 1 type/brand of caps . This you guys just have to experiment.
Cheers & Happy Diy
Thanks, nice tips 🙂 I may have to try M Lytics. How much capacitance there should be per channel? I have 30v trafos so CRCRC-config maybe to reduce ripple and lower voltage a little?
You are saying that the electrolytic caps doing ripple filtering in the PSU has an effect on sound quality transparency and forwardness? Perhaps on the Aleph feedback cap and main feedback shunt capacitor as those are in audio path. But I think the power supply ripple filtering caps don't really play a part in the audio signal path, except from standpoint of ensuring low noise floor and giving some reserve for bass transients. But this is Class A where bias current is fixed. I think lower ESR power supply caps will give better bass authority, not sure how it affects the mids and highs and transparency or forwardness though?
Apologies I not familiar with this amp. Can’t recommend the carrect values to used. My own findings is never to over use capacitance cause it will make the amp sound too tight & unnatural
Cheers
Cheers
zman01,
The transformers are RS PRO 500va 30v rebranded Nuvotem Talema made in the Czech Republic. $84.93 AU each.
X, Caps are Kemet 15000uf 63v from Mouser I reckon these are a bargain at $7.25 AU each but i did buy 16 so it does add up but i'm only going to build the BB Alpha one time At least i hope so.
Steve.
The transformers are RS PRO 500va 30v rebranded Nuvotem Talema made in the Czech Republic. $84.93 AU each.
X, Caps are Kemet 15000uf 63v from Mouser I reckon these are a bargain at $7.25 AU each but i did buy 16 so it does add up but i'm only going to build the BB Alpha one time At least i hope so.
Steve.
Hi XRK
Proof of pudding is to try it for yourself. Things like these have been argued for as long as I care to remember. Im sure Hgh Dean would agree with what I say
Cheers
Proof of pudding is to try it for yourself. Things like these have been argued for as long as I care to remember. Im sure Hgh Dean would agree with what I say
Cheers
Could a good guess be speed of available current when amp needs juice is proportional to actual AC impedance of last stage of capacitor bank that amp can see as its feeding rails.
Below is a homemade AC impedance overlay for 1uF/10uF/100uF/220uF/470uF/1mF/2,2mF/6,8mF/12mF/35F, regarding dielectric 1uF and 10uF stand out in they PP film where all the rest are traditional electrolytes, the big value 35F one is not very practical in its voltage range is a low 2,5V number.
Think one could kind of scientific investigate this point measure a power amp using low value power bank verse same amp using big value capacitor power bank, then in REW look for if excess phase at lows increase some cycles for the low value power bank verse the big value one, kind of proportional to how impedance graph for the two between will look.
Attachments
Last edited:
Thank,s Byrtt. I can't tell which curves are which as there is no legend. Do they all propagate right to left with the ordering you gave above? Which lines are the polyfilm ones (first two that are straight lines)?
Sorry i kind of thought overlay would give itself and some of the reason why we pick high or low values for a particular job whether its for filter shunts or coupling : ) the right order of traces is higher values starts at left so the two strait lines at right are the smaller values of 10uF and 1uF that also happens to be the PP dielectrics, in reality those strait lines would look close to scheme of the bigger values its just that it happens up at a frequency not covered by DATS bandwidth.
Juntuin,
Those cpu coolers will work just fine. They are unbelievably efficient.
Nice massive trafo. 😀
Great to see so many B.B. builds starting to gear up.
Those cpu coolers will work just fine. They are unbelievably efficient.
Nice massive trafo. 😀
Great to see so many B.B. builds starting to gear up.