B251 Shortness of breath?

[Wally John]

Dear friends,

It's been almost a year since I fully restored my last B251. I restored 3 total yet but I must say the last one was the holygrail of all. Don't get me wrong, it is still functioning very well and this the one that is sounding the best of all. BUT!

I don't know if Milan is still around on this forum but I 100% agree with something Milan Milosevic wrote into another forum in 2022 in regards of the B251.

Here is his statement:

"Properly designed switching power supplies have lower noise and lower ripple, because they operate at higher frequencies. The audio industry welcomed SMPS because it allowed for the design of high gain stages with very low noise, as in Sony TA-F6. They also provide excellent transient response, as in Revox B251 (2μS rise time). However, they also exhibit some "shortness of breath", and they cannot deliver the same bass punch as their linear counterparts: these amplifiers will always sound a bit "light", while they excel in ranges above. In general, SMPS are best suited for constant loads.
All other cons of SMPS, such as noise and EMF pollution going back to the mains, are due to cheap and poorly designed circuits."

It is not my nature to give-up easily and wondering if this "problem" can be fixed, improved or just there is nothing we can do about it.

If I'm comparing with other amplifier designs, I'm often seeing large reservoir capacitors (thousands of Uf) on the voltage rail feeding the amplifier output stage. However, for the B251, the only candidates for this task would be the C113, C113, C312 and C313 of value of 220uf only. I'm wondering if increasing the capacitance to ~ 1000uF would be beneficial at all or maybe detrimental because to current surge added onto the power supply to "fill" those?

Let me know what you think!

Thank you!

Charles

[schori]

Note that there is not just one 220uF capacitor in the poweramp. There are also 2 such capacitors in the power supply unit. Together with the 40uH coil, they form a low-pass filter for the 22kHz switching frequency of the power supply unit. So if you want to have more energy reserves, the power supply unit must also be able to transfer this energy. So the primary side with its C25 must also have this energy available, and the rectifier and the coil in the low-pass filter must also be able to deal with it. The same applies to the main transformer.

Without knowing exactly, I suspect that this doesn't really make a difference. But since everything is built symmetrically, you can try it out on one channel and see if one channel sounds better than the other.

And then of course you have to ask yourself why Revox didn't do this....are there any reasons? (EMC, heat, lifetime of parts.....)

I'm looking forward to your experiments...

Stephan

B251_powerSupply.png

[Wally John]

Hello Stephan,

Thank you for your kind input.

I was also suspecting that increasing the amp output capacitance would require beefing up the power supply. But like you said, doing that is would involve changes all the way to the power supply primary... within the available space I have.

The other question is: Why Revox didn't do this? Well, not to offend anyone about the brand but I really think Revox dropped balls with the B251, i.e. Horrific thermal management, cheap PCB's, borderline component selection...

All in all, I might just go half way and replacing capacitors with value of 470uF instead and see what it does. Strangely, Revox didn't implement the NTC surge protection circuit for the US (120V I have) version (schematic at the end of the service manual) so the components are taking the hit for the inrush current...

I will do the change and report back if it did improve the bass punch... not having too much hopes though...

Regards,

Charles

[MilanMilosevic]

Hello John,

The reason for considerably smaller capacitors in B251 PSU is because they charge at higher frequency (22kHz) and therefore faster. In other words, their duty cycle is shorter. In conventional power supplies, PSU capacitors are charged in slower 100 or 120 Hz pulses and require a larger reserve. Of course you can increase the capacity from 220 uF (do not overrate more than 50%), but that could have adverse effect on sound, including worse transient response. However, the "lightness" is inherent to SMPS as such and, so, the B251 for me remains the most beautifully sounding Revox amplifier but it will never have the bass punch of B250.

[Wally John]

Thanks Milan for taking the replying to my message. I'm now happy with my B251 and I didn't change a part! After a year of its restoration, I revisited the bias and I don't know why but it was way too low. I've used the same adjustment technic but it made a huge difference this time. I can only blame the aging of the new components that affected the bias eventually. I'm now at 10mV and I'm getting a nice bass even at low volume. I'm super happy now Regards,