B1 Buffer Preamp

[Babowana]

Can anyone recommend a good potentiometer for the B1?

I'd like to recommend any audio-grade pot within your budget.

 

[schro20]

BSC circuit + vol control with B1? How?

Dear Fellow DIYers,

I intend to experiment with line level BSC correction along the lines of http://sound.westhost.com/bafflestep.htm. I am thinking that it makes sense to do this in conjunction with a B1 and while I am at it, I want to integrate the overall volume control in this manner as well (for that matter I want to use the other ESP suggestion of pseudo log'ifying a linear pot with a resistor from wiper to ground http://sound.westhost.com/project01.htm).

I am less certain what it means to cascade the volume and BSC controls. Regarding the BSC circuit of Rod Elliott:

An externally hosted image should be here but it was not working when we last tested it.

he says:

It is essential that the compensation circuit be driven from a low impedance source, and the load impedance should be reasonably high. There will be little error with loading above 20k, but basically the higher the impedance, the better. Opamp buffers at the input and output may be used if you cannot ensure that the source impedance is 100 ohms or less, and that the load impedance of the following stage is greater than 20k. My recommendation would be to use a buffer stage at the output with an input impedance of about 100k.

What does this imply regarding the combination with a B1 buffer and a volume control? If I assume 100k linear pot and 16k for the wiper to ground resistor I get approximately 10k impedance at wiper mid position (16k||(50k||50k)). To ensure a low source impedance for the BSC it should be preceded with a B1 I presume. The requirement for the high load impedance of course suggests that I should have a B1 after as well. The volume control would not be enough load impedance to go right after the BSC circuit, so it should go in front: Vol/B1/BSC/B1

Am I getting this right? Any suggestions of how to accomplish the BSC and vol control circuit functions together without going overboard as far as buffers are concerned?

Thanks much for your input!

peter

 

[soundengine355]

does the B1 suffer from impedence mismatch? Ive read that most passive preamplifiers have this issue.

 

[Nelson Pass]

"It is essential that the compensation circuit be driven from a low impedance source, and the load impedance should be reasonably high"

I don't think that it is essential at all. If you are not looking
for compensation, then remove the circuit altogether.

does the B1 suffer from impedence mismatch? Ive read that most passive preamplifiers have this issue.

Isn't matching impedance in a passive the whole point of the
B1 in the first place?

 

[schro20]

"It is essential that the compensation circuit be driven from a low impedance source, and the load impedance should be reasonably high"

I don't think that it is essential at all. If you are not looking
for compensation, then remove the circuit altogether.

But I do want the BSC circuit in there. I want an adjustable BSC at line level. Maybe I should be talked out of that, but given that I want it in there, are you suggesting that a single B1 buffer stage is enough? Say, after vol and BSC circuit?

Peter

 

[woodturner-fran]

I'm just about to put my B1 together with a lightspeed attenuator in place of the normal pot for volume control. Can someone just double check me here? (I will have a separate switch for source selection because I have 3 and maybe 4 sources.)

I would normally just go from input switch directly to the pot/lightspeed and from there to the PCB input (ie output and GND from the lighspeed to W and CCW on board), but in this case R100/101/200/201 are in the mix. Is it ok to do it this way?

Or should I route the signal through the board (inputs from RCAs to switch, outputs from switch fed to middle set of pads on the board (ie the ones not marked L1, L2, R1, R2) Then connect CW to the input on the lightpseed, W to the output from the lightspeed and CCW connected to GND on lightspeed.) It seems theres no advantage in doing this?


What do you think?

Fran

 

[Nelson Pass]

But I do want the BSC circuit in there. I want an adjustable BSC at line level. Maybe I should be talked out of that, but given that I want it in there, are you suggesting that a single B1 buffer stage is enough? Say, after vol and BSC circuit?

It may easily be that no buffer is required. If your source
impedance is reasonably low, say 1K, and the load is high,
say 47+K, then I think the filter will work find as is.

This presumes that both source and load don't have issues
with the impedances they see, but I certainly would try
it first before worrying about buffers.

 

[budwiser]

So will any of these power supplies work?

http://www.mouser.com/Power/Plug-In-AC-Adapters/_/N-wp53?P=1z0wdj3Z1z0y701Z1z0wd9nZ1z0wd93Z1z0y6z2

 

[woodturner-fran]

Thought I'd post a few pics. Only got her finished tonight....

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

I'm using a tread PS with some extra filtering from a wallwart putting out 18V AC. I used a 50K stepped attenuator. I originally planned to use a lightpeed in there, but that didn't work out (think the pot I had wasn't right or something). Have ot investigate that more.

Case is an cast alu job repurposed from an old PA speaker. Kind you would have seen in a church or hall 30 years ago.

Fran

 

[Skorpio]

the B1 that Pass has designed runs the amplifying FET at very close to Idss.
The lower FET is a CCS to keep the upper FET at or just below Idss.
The two FETs are doing different jobs.
They do not need to be matched. They need to be selected for similar or equal Idss.
As Nelson says use the lower Idss as the lower device (CCS). One does not want the signal modulation to take the amp FET current too far above the Idss current.

This is not consistent with post #148 where is is stated/tested:

Conclusions:
1] 2N3819 need at least 10V more Vsupply, but then will work very well
2] 2nd harmonic is effectively cancelled, at higher output voltages, 3rd harmonic dominates.
3]it's best to keep the higher Idss JFET as the CCS.

 

[Nelson Pass]

Probably that would be with the 2N3819.

 

[Nelson Pass]

I just noticed this:

did Nelson make a mistake when he write in Side E "A teensy bit of second harmonic" .. I would think he means 3rd harmonic..

Sort of. Much of the time the distortion is 2nd, but at the
point where that shot was taken, the circuit was enjoying
what Scott Wurcer calls "the fortuitous null", leaving only
the 3rd. This is the result of load line cancellation at a
specific voltage and current, and occurs at one spot for a
given load value.

 

[rgheck]

Can anyone point me to a design for a regulated power supply for the B1? I'd prefer something that will drive both halves of a balanced version. There's probably a design or two in this thread somewhere, but it's gotten kind of long!

 

[AndrewT]

This is not consistent with post #148 where is is stated/tested:
..........................
3]it's best to keep the higher Idss JFET as the CCS.

Hi,
I cannot agree with that conclusion irrespective of what the measurement shows.
Mr Pass suggests the higher Idss be used for the amplifying device.
That Mr Curl points out the deficiencies if a jFET is biased above it's Idss
I cannot recall seeing further measurements with a load as suggested in a later post.

 

[tschrama]

Hi,
I cannot agree with that conclusion irrespective of what the measurement shows.
Mr Pass suggests the higher Idss be used for the amplifying device.
That Mr Curl points out the deficiencies if a jFET is biased above it's Idss
I cannot recall seeing further measurements with a load as suggested in a later post.

Hi all,

Nice to see this buffer schemtic still being discussed. As it turn out, Just yesterday I started to build another B1, this time from a PCB i got from feebay.

Reading up on all the discussion and my own measurements, I again got intersted in this discussion about which JFET, if slightly unmatched, should be the CCS....

Too bay my distrotion analyzer got fried.. so here's what I did. From basic Gm and Idss measurements I fine tuned my spice models to precisely match my 2Sk170s. Again , but this time from simulations only on a 10K load, for lowest THDm the higher IDSS should be the CCS.... why? I don't know, but it confirms my measurement from many moons ago..

I guess, the Vgs might be slightly possitive, but that doesn't hurt as long as there is no significant variation of Vgs, which incase of the CCS loaded folower into 10K is the case.. Being slight baised positive, the JFET enjoyes a higher Gm at a slightly highr Iq.. this might bring teh THD slightly down I guess..


PS
Anybody got some tips on a nice dual taper audio pot for the B1?

Kindest regards,
Thijs Schrama

 

[AndrewT]

Hi T,
thanks for the follow up.
Although it is opposite to what I want to believe, it's nice to know that it may work to our advantage.
Who wants to experiment by swapping the jFETs around and tell us what they hear?

 

[Skorpio]

Input resistance versus volume resistance...

I am about to order a relay controlled attenuator for B1. I is possible to get the following resistances: 1k, 5k, 10k, 20k, 50k and 100k. The input resistance is constant regardless of damping...noting about output resistance...

B1 input is 1M, but kan easyly be lowered is this might be better?

What is the optimum choice and what are the pros and cons?

(my first impulse was to order 100k to load source as less as possible...)

 

[AndrewT]

A major consideration when selecting a source impedance/resistance is what follows it.
If the following cables and/or receiver have significant capacitance then the source impedance (Rs) must be selected to allow the required frequencies to pass through.

A 100k pot has a maximum output impedance of [100k + Rs] / 4 ~ 25k.
This 25k will interact with any following capacitance to form a Low Pass filter.
Most require the Low Pass filter to have F-3dB >=50kHz and some ask for >=300kHz.
For 50kHz the capacitance must be <= 1/ 2/Pi/50000/25k <= 127pF

The pot is followed by a buffer. The parasitic capacitance of the PCB + the capacitance of the connecting cable + input capacitance of the buffer <=127pF for F-3dB >=50kHz.

If the B1 is connected to the 100k pot with a tisted pair about 50mm long there is a very good chance the total capacitance after the pot <<127pF.

If you find that the treble with the 100k pot is slightly rolled off, then a lower impedance pot can restore the lost treble by moving the F-3dB higher up out of the audio frequency range.

A good choice for a pot that can transmit into any nearby circuit is about 10k.
That's what I would suggest.

 

[Melon Head]

I thought the idea of this was so that you could use a pot of higher while still maintaining a relatively low output impedance.
If you use a 10k pot then the benefits of having a buffer are reduced or made redundant.

 

[AndrewT]

Hi T,
go and read my reply again.
I did not say you cannot select a 100k pot + buffer to drive cables to a remote amplifier.