[FS] DIYAudio First Watt/Sony VFET Amplifier

[Dan K]

Good Evening All,

Dare one ask why???

Here goes..
To properly show their character the VFETS need +/- 28V, but will 'work' with anything from 26V to 30.
These amps can be fiddled to make sound with a 24V PSU (at least according to NP), but the results are nowhere near the/ his design intent.
The front end of these is a mildly reworked F5 and needs 1.5V of bias (anything less and its not going to properly drive the VFETS, anything more and it will start to get quite hot)
The CCS' for the VFETs need 2V across their regs , or these will drop out = sub optimal performance
The VFETS need 125+ mV of bias to reveal their 2nd/ 3rd/ 4th order distortion profile and sound like a triode
Maybe with many resistor value changes a sweet spot can be found where the above will all work with 24V.
Massively easier to fit new torriods that put out the correct voltage

...and then spend every weekday evening for the past 6 weeks tuning the thing to get the desired distortion / delicacy.

You've got to put the hours in tweaking the bias until the VFETS spring to life, and then find the spot where they reveal their character with the lowest achievable noise floor, it's a bit like tuning an old FM radio.

Actually quite far from easy, but it does now sound like an F5 that's had its output stage replaced with a pair of EL34

 

[k90tour]

I wondering (out loud) how it would compare to my Accuphase A20.

 

[Dan K]

I wondering (out loud) how it would compare to my Accuphase A20.

The A20 looks to be an integrated.

The R0 will put 15W into 8 Ohms with a strong active pre behind it. You'll need a really good pre and efficient speakers. Something like:

Any of the Concordants
Audio Research LS25 & upwards
Conrad-Johnson?

Speakers will likely be:
Monitor Gold
NS1000M
H2
Klipsch Horns

This amp is for someone who adores/ needs a valve amp, but for domestic reasons isn't allowed one. It would be interesting to trial it against a Radford STA15. It should be quite similar, but possibly with a lower noise floor - and Radfords have extremely low noise!

 

[k90tour]

The A20 is a Class A power amp.
I do like valves though.

 

[Dan K]

The A20 is a Class A power amp.
I do like valves though.

What are your speakers?
I reckon this amp would be wonderful into a pair of NS1000Ms, it's very good into Snell J's. My problem is I have a fetish for Omnis and it's not really able to get my Sonabs going.

 

[k90tour]

Harbeth HL-P3ES

 

[Dan K]

Harbeth HL-P3ES

"Impedance is complex, but nominally 6 ohms (treat it as 4 with tube amplifiers). Sensitivity remains a low 83dB, recommended minimum power 25 Watts"

6 Ohms is good, but 83dB? Maybe others with specific Harbeth Knowledge can advise.

 

[irb]

Here goes..
To properly show their character the VFETS need +/- 28V, but will 'work' with anything from 26V to 30. ...

Actually quite far from easy, but it does now sound like an F5 that's had its output stage replaced with a pair of EL34

Makes sense now.

Incidentally, I initially read your ebay name as dank night 1000, which I rather like.

 

[Mike P]

Incidentally, I initially read your ebay name as dank night 1000, which I rather like.

I did that too, before the penny dropped.

 

[Dan K]

I've one more tweak that I may have to try. The Rails are 1/2 a volt above nominal at 28.5. I could up the R value in the CRC and get a small further reduction in PS noise/ ripple, although I do already have 0.4 and 0.5mV. Very unlikely to be audible, but who knows. Problem then is having to reset all the parameters mentioned. Otherwise the toolbox is empty

 

[Dan K]

Decided to bump this, as despite over 500 views and 47 watching on eBay it remains with me.

I've now replaced 3-off 47k shunt resistors on each board with Kiwames, which did result in a further improvement/ 'easyness'.

I'm now waiting on a further HFC order to replace all 4 of the gate stoppers, as well as the feedback series resistors with AN Niobiums

I have too many amps

Let me know if you're interested.