Here is almost completed integrated amp, which unifies No NFB headphone amp/preamp with the Modified Follower. As expected, sounds exellently.
Output "cap" is a ring-like cluster of 1000uF low-ESR caps (SamWha WL) with 100uF polypropylene shunt (Solen).
Current was decreased to 2A, since with 60V power rail power dissipation goes to high for a single transistor.
Output "cap" is a ring-like cluster of 1000uF low-ESR caps (SamWha WL) with 100uF polypropylene shunt (Solen).
Current was decreased to 2A, since with 60V power rail power dissipation goes to high for a single transistor.
Attachments
Member
Joined 2009
Paid Member
Looks very nice Vldimirk.
I suspect you must own a heatsink company
Some years ago I was inspired by Nelson Pass for building Zen4 and Zen9, then I wanted more and more power from them, and this has ended up by ordering of 6 big custom made (machined) heatsinks. Since that time, only Zen9 (4,8A current, disconnected CS modulation) is living still, another two pairs of heatsinks have travelled to follower-based designs.
Now the Son of Follower is being breaking in. Pass's style has touched a bit this creation.
I would also like to mention, that VT9 and VT10 can be easily replaced by more obtanible MOSFETs, since output impedance of the driver stage is nearly 2 Ohms (it is possibly a record value). I used to connect the driver stage directly to loudspeakers, and sound was quite listenable.
I would also like to mention, that VT9 and VT10 can be easily replaced by more obtanible MOSFETs, since output impedance of the driver stage is nearly 2 Ohms (it is possibly a record value). I used to connect the driver stage directly to loudspeakers, and sound was quite listenable.
Attachments
Last edited:
Member
Joined 2009
Paid Member
Pass's style has touched a bit this creation.
Indeed, I noticed also the front panel design
PCB
A possible PCB for the power amp.
The board is not include the large capacitors and two resistors: C6,7, 11, and 12, the resistors R20, 23 and R1 the potentiometer.
The size is 17.8x4.5 cm (7x1.8 inch).
A possible PCB for the power amp.
The board is not include the large capacitors and two resistors: C6,7, 11, and 12, the resistors R20, 23 and R1 the potentiometer.
The size is 17.8x4.5 cm (7x1.8 inch).
Attachments
Since getting HF transistors is not an easy task, I would propose using K1530 as VT9, VT10. At this, R15 and C4 could be decreased for preserving the passband of the amp.
Also, now I can state that the Low-ESR SamWha WL 1000uFx63V, in the cluster arrangement, give very good result for the output cap. Of course, one could test also Nichikon KZ or Elna Silmic II, for instance. With 100uF polypropylene shunt, sound easily competes and outperforms that of a good tube SE amp, at all frequencies.
Also, now I can state that the Low-ESR SamWha WL 1000uFx63V, in the cluster arrangement, give very good result for the output cap. Of course, one could test also Nichikon KZ or Elna Silmic II, for instance. With 100uF polypropylene shunt, sound easily competes and outperforms that of a good tube SE amp, at all frequencies.
Member
Joined 2009
Paid Member
Vladimirk,
You know, it's kind of strange to see that you are from a city world famous for the quality of it's vacuum tubes, yet you are a solid state guy ?
My amp building activity is not aimed against tubes, simply I believe that HF transistors and tube-like simple SS schematics could do the job even better (sound-wise). But, for achieving this, one should observe few design principles. One of them - make the regions with alternating currents as compact as possible, for precluding parasitic interactions, induced noices and post-effects. Shunt PS and follower-like solutions are very good for this.
Many "perfect" schematics with very low THD do not sound good, since the sound is sensitive to many usually unaccounted things.
Testing of the Son of Follower goes on, the reference is a high-end tube 300B SE amp from NEM (Novosibirsk), also integrated amp.
The tube amp needed nearly 200 hours for burning in, while the S of F - only 20 hours. There is no frequency range, which is better performed by the tube amp.
The tube amp needed nearly 200 hours for burning in, while the S of F - only 20 hours. There is no frequency range, which is better performed by the tube amp.
Attachments
Last edited:
Testing of the Son of Follower goes on, the reference is a high-end tube 300B SE amp from NEM (Novosibirsk), also integrated amp.
The tube amp needed nearly 200 hours for burning in, while the S of F - only 20 hours. There is no frequency range, which is better performed by the tube amp.
Looking good.
Seems the 300B amp as big as your Son of Follower.
Why the difference in the burning in times?
Looking good.
Seems the 300B amp as big as your Son of Follower.
Why the difference in the burning in times?
The 300B amp belongs to the family that got several prises at Las Vegas, High-End cathegory. Its weight is 65kg, and output trafos are big and designed for 4 Ohms load. Also, PS includes vacuum rectifier, big choke and Black Gate caps. Output impedance is 0,6 Ohms.
Long burning-in time, I guess, is usual for trafo-based devices. Capacitors and solder joints are much easier in this respect.
With the follower-like designs, by improving output cap, one can really achieve all that tube SE excellence.
Very impressive Vladimir. I my self stay away from audio stuff which I can't lift alone.
So my limit would be in the 20, max 25kg range.
Hello, Miklos
This weight, nearly 20kg, is just what I have for the Son of Follower, at this I do not account for the PS trafos, that are external devices. Actually, its weight could be around 12-15kg, with more optimal heatsinks. My heatsinks are overspeced for this application (in this schematics we have 2A x 60V per heatsink, while these heatsinks can do 3,8A x 60V).
Hi Vladimir,
I tried the cap cluster on my amp, now i have 10 X 1500 uF RIFA caps per chanell and 2x33 uF polipropilens. The first impressions are dissapointing. Before, with SPRAGUE 4700uF per chanell sound was much better. Maybe I should give some time to RIFA's ?
I like how your son of follower looks. But i dont like the name you choose for it
Regards
Goran
I tried the cap cluster on my amp, now i have 10 X 1500 uF RIFA caps per chanell and 2x33 uF polipropilens. The first impressions are dissapointing. Before, with SPRAGUE 4700uF per chanell sound was much better. Maybe I should give some time to RIFA's ?
I like how your son of follower looks. But i dont like the name you choose for it
Regards
Goran
Hi Vladimir,
I tried the cap cluster on my amp, now i have 10 X 1500 uF RIFA caps per chanell and 2x33 uF polipropilens. The first impressions are dissapointing. Before, with SPRAGUE 4700uF per chanell sound was much better. Maybe I should give some time to RIFA's ?
I like how your son of follower looks. But i dont like the name you choose for it
Regards Goran
Hello, Goran
I could only confirm, that Elna Silmic, Elna Cerafine, SamWha WL, both measure at 10kHz (dielectric losses angle) and sound good.
I have not used RIFA, in spite that I bought 4 pcs of them (PEH 169 4700uF 63V) in Singapore and intended to use them, but after doing measuremens decided not to spoil time.
Not forget also, that any new assembled device needs 20 hours of intensive burning-in (I use white noise from FM radio, adjust level for having 6...7V AC voltage at the output).
What name for the amp would you prefer? I simply have not paid attention to this aspect. At least, Son is much better than Death (sometimes used).
Last edited:
Hello, Goran
I could only confirm, that Elna Silmic, Elna Cerafine, SamWha WL, both measure at 10kHz (dielectric losses angle) and sound good.
I have not used RIFA, in spite that I bought 4 pcs of them (PEH 169 4700uF 63V) in Singapore and intended to use them, but after doing measuremens decided not to spoil time.
Not forget also, that any new assembled device needs 20 hours of intensive burning-in (I use white noise from FM radio, adjust level for having 6...7V AC voltage at the output).
What name for the amp would you prefer? I simply have not paid attention to this aspect. At least, Son is much better than Death (sometimes used).
Hey Vladimir,
Yes, I think RIFA's need more time to burn in (if they ever will). I can generate white noise from cool edit, but i think wife will kill me after 15 min
Abt name, i would like you use a name of something whats close to You, even written on cirilic, it will be cool, the design is yours.
Regards
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Solid State
- Integrated No NFB Son of Follower