B1 Buffer Preamp

pcb mods

Uriah,
If you were to move the C100,200 apart and shift R4 to one side of the Q100,101 so then can move the four Q's together (after turning the Q200,201 around) to get better tracking - just a small thing, but ...
Also, if you were to move the R102,202 gown closer to the Q's, then can shorten the pcb.
Alternatively, can add a simple zen type reg or a Cmultiplier to input power rail.
 
jh,
Thanks for your response and time.
I spent maybe 15 minutes trying to change the Q's around. The traces dont want to behave and let the leads of the Q's be accessible. If this were a 2 sided board it would be a cinch. I tried, instead of putting Q101 and Q201 face to face, to put Q101 and Q100 face to face but the same sort of problems pop up with traces needing two sides.
Making the board shorter will not help in this case as I am using some giant poly caps for C101 and C201. They are really what is restricting the size of the board. I would like to compare the sound of the large polys with many other 10uf caps because this board could really be tiny with the large electrolytics standing upright and the rest of the circuit pushed together a bit. Also if the 1ufs were tantalum or ceramic, but thats an experiment away.
Could you tell me about the cmultiplier and why I might want to use this?
Thanks jh,
Uriah
 
These simple ccts are quite dependent on the quality of the supply -(better power supply, better sound) - can use smaller high quality caps like those poly's with a Cmx chosen for sound characteristics - few parts, big difference.
The humble little Cmx cct often will give a more pleasant result than some of the more sophisticated regulators, and so easy to try out. Suggest you give it a go - plenty room on pcb!!
 
B1 born in the land of Oz

Firstly, thanks to Nelson for this simple design and to Salas for his work on the modified capless split rail version. The one I built is based around the schematic that Salas provided.

Very easy build, doesn't come much more simple than this, and no problems experienced. The circuit is extremely quiet with no associated nasties. I wasn't too sure about no caps even though the previous buffer (BB) in this pre also had none, so I made the board large enough to allow for caps if required.

Upon firing up, the offsets were -3.1mV, +0.2mV and were adjusted to -2.7mV, +0.7mV via the trimpots on the power supply. With these offsets the power supply was doing ±10.1V and never varied more than 0.04V during all the testing. Changing the offsets to equal values did cause the rails on the power supply to differ too much. The offsets and only varied 0.2mV during hours of testing and playing, so no caps were used. Matched pairs of 2SK170BL were used (thanks for them Kovax) and maybe would have better offset results if 4 were matched.... dunno.... but the offsets I have is fine.

The initial tests and listening was done with a 10K Alps RK27 as that's what was in the case and the sound was as expected.... very transparent, open, smooth, good dynamics and wide range, in other words, much better than a standard passive but with the good attributes that a passive can provide. The 10K pot was then changed over to a 20K Alps RK27 and the sound was not quite as good. Some of the dynamics were lost, bass seemed more recessed and not quite as open. So the 10K pot went back in as well as the wiper shunt resistor changed to 100K from 220K. This is where it has stayed. I have no idea if the sonic variation was due to the quality of the pots, but both are Alps RK27 motorised and about the same age.

It was tested in 3 systems and 2/3 performed extremely well with one not reaching the full potential on offer. I think the simpler the system, the better the performance with the B1 and has ended up in my office partnered with an AudioSector LM4780 (reduced gain) and some Vifa based nearfield monitors. The B1 is just what the doctor ordered in this system and has taken it to a new level.... plus no additional gain.... woohoo. For those with ChipAmps and simple speakers such as fullrange, I think the B1 could be the ticket to better sonics. The other system with great performance used a Anthony Holton NXV200 and a stand mounted 2-way using Vifa XT25TG / Scan Speak 18W8531... the NXV200 never sounded better.

There will be more of these made for sure as I can think of a couple of other uses for the B1 (and no GAIN.... great).
 

Attachments

  • rzb1_case.jpg
    rzb1_case.jpg
    91.4 KB · Views: 2,612
Re: B1 born in the land of Oz

rabbitz said:
............ Changing the offsets to equal values did cause the rails on the power supply to differ too much. The offsets and only varied 0.2mV during hours of testing and playing, so no caps were used........

I think you mean did NOT? might not be obvious to non-native(?) English speakers ;)
 
Re: Re: B1 born in the land of Oz

pinkfloyd4ever said:


I think you mean did NOT? might not be obvious to non-native(?) English speakers ;)

I meant "did". You had me going there as I'm partially disabled and tend to hit the wrong keys or type slower than the brain so I miss out on key words.

When I adjusted the offsets to about the same numerically (1.4mV), even though one was +ve and the other -ve, the rails were quite different. I'm trying to remember but it was something like +10.1V and -9.4V so I decided to keep the rail voltages the same and have and extra 1.3mV offset which is no problem.

Salas

The next one I build I'll try and get them closer, but hey, a few mV here or there is not going to upset me.
 
It might not be a great idea because it may output much more than 20V under the tiny load that the B1 presents. You could use two 9V batteries or buy a 3 dollar wall wart on Digikey.
Also, you could buy a LM317 and follow the voltage regulator diagram on the datasheet for it. Hook it up to the output of your 20V supply and regulate it at 18V. The datasheet shows exactly how to set it to 18V.
Uriah