There is probably an improvement in reliability and ability to push higher voltage, hence dissipation on 20W. Maybe it can be a 25W or 30W amp with boosted rails and better heatsinks (CPU coolers with fans). Hugh can answer.
But the outputs are pricey so may as well make the 52W version!
I did not make BOM but you use regular BOM and use Vishay-Dale 0.5W 1% CFM55 resistors everywhere except Pan ERX 3W on R131/132. Also use all Silmic II 50V caps for electrolytics, and Wima FKP/MKP on all bypass. I went with 2.2uF on electrolytics. Then use your favorite 1uF to 10uF (depending on how deep you want your bass to go before rolling off) MKP 400V axial boutique cap on the input. That’s basically the formula. And good brand silver mica 500v on the two 18pF caps or good brand name 500v NP0 SMT works too.
Last edited:
I chose IXYS because they are available in TO264 for less than $20 each. IRFs may work fine, but I have not tried them. I've used a lot of IXYS devices and never had a problem. IRF was purchased by Vishay some years ago and their cost advantage has disappeared over time. They always were very good devices, but I have not tried them in the >500W ratings and cannot tell you the prices these days!!
Hugh
Hugh
Thanks X. I will probably be using the Alpha20's to drive the midranges of my active setup, so not needing high power and not a complex load, but the drivers do dip to around 5ohms. So a bit of current may be needed, hence wondering if the BB devices might be a bit more durable (and if there is any reason to think they might sound better). Your premium substitutes are as I suspected - will go down that path.
Pass Laboratories XA25 power amplifier Measurements | Stereophile.com
Aksa Lender P-mos Hybrid Aleph (ALPHA) Amplifier
Fortunately, the distortion in the midrange was predominantly the relatively innocuous third harmonic (fig.7), though at low frequencies the second harmonic was equal in level to the third (fig.. But note the very low level of all distortion harmonics in this graph!
It does indeed eek out a bit lower distortion with Vishay Dale resistors and all Wima and Silmic caps, now about under 0.006%THD for 2.75vrms into 8ohms:
The essential issue in the AKSA/Lender is the profile of the harmonic, which is close to monotonic decrease, in the classic approach identified by Jean Hiraga in the sixties.
Thank you for the graphs, Maty!
I do not regard THD per se the big issue. After all, a wonderful SE triode amp is up to 2-3% and no one sees this as cognitive dissonance.... It is the harmonic profile; this is what distinguishes the instruments of the orchestra, after all, and it is so with audio amplifiers.
However, I concede these ideas are very, very expensive to prove definitively. Unless there are strict procedures in academic engineering labs this work is not really practical for a manufacturer unless it has millions of R&D to throw at it. Speaking from my POV, it would be far more profitable to spend the money on the marketing, with a nice tract of advert copy and a very pretty girl draped across the enclosure....... After all, the shareholders would prefer ROI rather than incontrovertible audio proof!
HD
Thank you for the graphs, Maty!
I do not regard THD per se the big issue. After all, a wonderful SE triode amp is up to 2-3% and no one sees this as cognitive dissonance.... It is the harmonic profile; this is what distinguishes the instruments of the orchestra, after all, and it is so with audio amplifiers.
However, I concede these ideas are very, very expensive to prove definitively. Unless there are strict procedures in academic engineering labs this work is not really practical for a manufacturer unless it has millions of R&D to throw at it. Speaking from my POV, it would be far more profitable to spend the money on the marketing, with a nice tract of advert copy and a very pretty girl draped across the enclosure....... After all, the shareholders would prefer ROI rather than incontrovertible audio proof!
HD
Last edited:
Hi Maty,
Should I add you to the Alpha 20 GB2 list?
All,
We have 10 sets now and I can go to production. Please PM me your PayPal address along with country it will be shipped to. Price is $28/pair and shipping is $5 in US, $10 in Canada, $15 everywhere else.
Thanks,
X
I should start an Alpha GB thread...
Should I add you to the Alpha 20 GB2 list?
All,
We have 10 sets now and I can go to production. Please PM me your PayPal address along with country it will be shipped to. Price is $28/pair and shipping is $5 in US, $10 in Canada, $15 everywhere else.
Thanks,
X
I should start an Alpha GB thread...
Last edited:
I have a problem: I LOVE my cheap tweaked KEF Q100 coaxial speakers. From weeks ago I have an incredible sound after I have reached the limit in cleaning or almost of the DC, ripple, noise an interferences on my house. With very good vinyl rips mastered with tubes I get excited when listening to them. Before, it happen less often!
My hypothesis is that the masters made with tubes acquire H2 predominance. But without measures it is only an hypothesis.
kef q100 class a amplifier - Google Search nothing
My hypothesis is that the masters made with tubes acquire H2 predominance. But without measures it is only an hypothesis.
kef q100 class a amplifier - Google Search nothing
My problem: 86 dB and 4.7 Ohms minimum
KEF Q100: Laborbericht KEF Q 100 Seite 5 - i-fidelity.net
Probably the real sensitivity is more near of 84dB. Why? KEF LS50 announces 85 dB but the measure is 83 dB and 3.3 Ohms minimum.
KEF LS50
[Polish Review, KEF LS50] Zespoły głośnikowe - testy
to English with Google:
Traductor de Google
KEF Q100: Laborbericht KEF Q 100 Seite 5 - i-fidelity.net
Probably the real sensitivity is more near of 84dB. Why? KEF LS50 announces 85 dB but the measure is 83 dB and 3.3 Ohms minimum.
KEF LS50
[Polish Review, KEF LS50] Zespoły głośnikowe - testy
to English with Google:
Traductor de Google
If you drop supplies to +/-18v and boost current to 2.7amps that might drive 4ohms speaker nicely. I will confirm with a simulation. Will need to switch to IXYS outputs though for the 50w dissipation per channel. Still reasonable use same CPU cooler heatsinks or larger passive radiator.
Edit: The sims show that it can work well into 3.6ohm load at +18v and -20v (asymmetric rails) and 2.8amps for about 48w and 55w dissipation on the two MOSFETs. Set R131 to 0.22R and R132 to 0.1R. Pretty good low distortion performance out to 21wrms for even 3.6ohm load.
Last edited:
Hi X, could you please add me to the Alpha 20 GB2 list for 1 set.
I have 3-way speakers with 200 Hz and 1.4-1.6 kHz crossovers and thought an answer in earlier posts to use one Alpha20 for mids and highs then realized I got miniDSP and not a passive crossover. Anyway, I will begin with midhorn amplifier replacement.
Thanks
I have 3-way speakers with 200 Hz and 1.4-1.6 kHz crossovers and thought an answer in earlier posts to use one Alpha20 for mids and highs then realized I got miniDSP and not a passive crossover. Anyway, I will begin with midhorn amplifier replacement.
Thanks
Thats good idea, I was just thinking about ordering them, but I don't need exactly 100pcs of those
I went ahead and ordered some TO264 alumina pads for my 20W build (and others), and will have plenty to share...but it'll take up to 3 weeks or so for them to get here.