Good luck with it and we will be happy to learn about your progress Terry.
the increase in output offset is almost certainly due to the increase in ambient temperature affecting the operating temperature of the jFETs.
Thermally coupling the pairs with a strip of aluminium or copper may give a small reduction in offset drift as the circuit warms up.
Strange, the original docs called out for 12V. Most of the units that were built in the beginning of this thread were all 12Vac. I didn't see one case where any of them complained about the sound.
I originally was attracted to this because I heard so much about how the B1 pre was preferred for letting the true sound of an amp shine through. Seems hot-rodding this pre may go against that philosophy. Maybe not, I shall soon see.
We were making a controlled low heat paradigm in the early ones so we kept the bias and voltage across conservative. And it worked fine. Higher current bias through it and higher voltage across it gave it some PSU quality and anti-mains irregularity advantages. At least the majority agreed. Does not mean that the initial conservative guides are not producing a very good DCB1. You may listen to both configs for yourself and decide of course.
Pass DIY Addict
Joined 2000
Paid Member
After sorting out an improperly wired volume control with the LDRs, everything is up and running and sounds great! I'm on my 12th or 13th disk by now and the improvements are really shining through!
I've been making some upgrades over the past few weeks that include Caddock MP821 resistors in the mid and tweeter networks for my speakers, extra reinforcement for the upright part of my open baffle speakers, an Audio GD DAC instead of the one in my Adcom GCD700 CD player, and now the DCB1 with LDRs in place of my Marantz AV8003 "pure direct" mode. What a difference they all make! Simply beautiful!
Salas: thanks for the tip about the jFet leg trick, I didn't realize this was necessary as bias is increased. I plan to run north of 1A when I build a proper chassis for it.
Andrew: thanks for the heads up on the offset, I'll give that a try in a little while. After fixing my boo boo on the volume control, I now have offset of 0.0mV and 2.1mV after warmup. Not too concerned, but if I can bring it down, I would like to do so.
For now, it's back to the music!
I've been making some upgrades over the past few weeks that include Caddock MP821 resistors in the mid and tweeter networks for my speakers, extra reinforcement for the upright part of my open baffle speakers, an Audio GD DAC instead of the one in my Adcom GCD700 CD player, and now the DCB1 with LDRs in place of my Marantz AV8003 "pure direct" mode. What a difference they all make! Simply beautiful!
Salas: thanks for the tip about the jFet leg trick, I didn't realize this was necessary as bias is increased. I plan to run north of 1A when I build a proper chassis for it.
Andrew: thanks for the heads up on the offset, I'll give that a try in a little while. After fixing my boo boo on the volume control, I now have offset of 0.0mV and 2.1mV after warmup. Not too concerned, but if I can bring it down, I would like to do so.
For now, it's back to the music!
P.S. The leg trick is to "tune" the Mez differently when deciding high hot-rod. As it stands and up to 200mA, I have planted a deliberate local CCS asymmetry that fixes the THD profile better. That one is restored by the "leg trick" and makes it like a new Hypno there. Which benefits from the high bias and the THD "fix" is not as useful then, plays better restored when too hot IMHO. Still, the differences are slight.
Pass DIY Addict
Joined 2000
Paid Member
Thanks for the additional details on the leg trick, Salas! I'm now up to 226mA (three 20R in parallel). Original sinks are holding at about 42c - a 20c rise over ambient. This is probably about as high as I will push it until I get proper sinks in a nice chassis.
I'm discovering more and more beauty in my old disks! I'm surprised just how many of them have such a nice center image - this only came through on the best of the disks with my Marantz preamp. Listening to Loreena McKennitt now - just beautiful!
I'm discovering more and more beauty in my old disks! I'm surprised just how many of them have such a nice center image - this only came through on the best of the disks with my Marantz preamp. Listening to Loreena McKennitt now - just beautiful!
Ok, so maybe I misunderstood what this design was intended to be. i got interested because although I have a few commercial preamps, they all impart their own "sound" on the music played through them. I enjoy, doing A/B sampling of the amps that I have built. While reading through some of the different amp build threads I often here the amp designer suggest that a preamp will often "color" the sound and thereby mask the true fidelity of the amp I am testing. A few times I have seen the B1 buffer suggested as a good alternative to a simple pot or a preamp. When I saw this DCB1 offered in the forum store, I was under the impression that it followed in the steps of the B1 Buffer. In the store docs there is a BOM which calls out many items that should be the builders choice but there is only one choice listed for the transformer specs. I have read through most of this thread, and still don't see where a 15-0-15Vav transformer was deemed superior or even the "better" transformer for a true sound. So I ask, is there anything gained sonically from raising the voltage, or does it just raise the gain? My whole direction was to get something true, not just another preamp with its own sound.
Thanks, Terry
Pass DIY Addict
Joined 2000
Paid Member
Hi Terry,
With just a pair of jFets, a fixed resistor, and a variable resistor in the signal path of the DCB1, it is probably about as free from color as you can get. The Toshiba jFets are well know for extreme linearity (no emphasis or de-emphasis at any particular frequency). I just got mine up and running and find it superior to a $2500 preamp/processor that I've been using for a few years now.
The 100VA transformer and higher bias point (bigger sinks and smaller resistance in the spot right next to the power supply caps) merely creates a more robust and stable power supply for the DCB1. What I've been learning over the past few months is that the better/cleaner/larger/more stable your power supply, the better the resulting sound quality. This doesn't impart any type of extra "sound" or "sonic signature," it just lets more of the original recording come through because you are keeping more of the power supply grunge at bay. The bigger transformer is worth it for the flexibility of future bias changes.
With just a pair of jFets, a fixed resistor, and a variable resistor in the signal path of the DCB1, it is probably about as free from color as you can get. The Toshiba jFets are well know for extreme linearity (no emphasis or de-emphasis at any particular frequency). I just got mine up and running and find it superior to a $2500 preamp/processor that I've been using for a few years now.
The 100VA transformer and higher bias point (bigger sinks and smaller resistance in the spot right next to the power supply caps) merely creates a more robust and stable power supply for the DCB1. What I've been learning over the past few months is that the better/cleaner/larger/more stable your power supply, the better the resulting sound quality. This doesn't impart any type of extra "sound" or "sonic signature," it just lets more of the original recording come through because you are keeping more of the power supply grunge at bay. The bigger transformer is worth it for the flexibility of future bias changes.
@Terry
There is no gain. Its a times 1 buffer. Its designed to be neutral. The lack of signal capacitors VS normal B1 was a fundamental move to augment that goal.
Various further enhancements promote just that. 12-0-12VAC Tx is just enough. More, solidifies the main constant current source a bit further and allows for better tolerance against local cases of weak mains around the world. Your choice.
There is no gain. Its a times 1 buffer. Its designed to be neutral. The lack of signal capacitors VS normal B1 was a fundamental move to augment that goal.
Various further enhancements promote just that. 12-0-12VAC Tx is just enough. More, solidifies the main constant current source a bit further and allows for better tolerance against local cases of weak mains around the world. Your choice.
Fair enough. Perhaps someone could edit the BOM offered in the Forum Store to reflect your findings. The price for the 15V is exactly the same as the 12V. Of course, now that the 12V Tx is on its way to me, I have another $30 to spend to get the preferred unit. So now the next question. Does the unit pull more than 50Va when operating at 15Vac or is this another case of more is better?
I understand the logic of a more solid PSU but I don't understand the desire to keep raising the bias if it is a neutral sound we are after.
Thanks, Terry
I understand the logic of a more solid PSU but I don't understand the desire to keep raising the bias if it is a neutral sound we are after.
Thanks, Terry
The higher transformer voltage is to prevent the regulator dropping out when voltages are low.
It has nothing to do with gain.
It is to stop the PSU glitches getting through to the buffering jFETS.
Both the B1 and the DCB1 run at a bias determined by the Idss of the lower of the two jFETs.
The PSU simply supplies electricity to these two jFETs (per channel).
Does it need more than 50Va? I suppose I need to order another Tx. I'd like to do that just one more time. If 50Va is sufficient, The Tx is smaller and lighter whcih may help when it come to building my case.
On second thought, I may wait until my Tx gets here and measure it's output. Many times, Antek transformers put out more voltage at my house due to the 60Hz and the 120V mains.
thanks
On second thought, I may wait until my Tx gets here and measure it's output. Many times, Antek transformers put out more voltage at my house due to the 60Hz and the 120V mains.
thanks
Last edited:
the PSU supplies ~10mA to each buffering pair.
The shunt Reg is fed by a CCS. That CCS current is set by a resistor. You have two of these.
A 50VA 15-0-15Vac transformer has a maximum continuous DC output of ~830mA.
For cool running use ~420mA as your maximum continuous DC feed to the CCS. The tapping for the relay draws ~10mA to 20mA from one polarity depending on how you wire it.
400mA is a lot of hotrodding for the DCB1. You can set it up for as little as 50mA if that is what you want.
The shunt Reg is fed by a CCS. That CCS current is set by a resistor. You have two of these.
A 50VA 15-0-15Vac transformer has a maximum continuous DC output of ~830mA.
For cool running use ~420mA as your maximum continuous DC feed to the CCS. The tapping for the relay draws ~10mA to 20mA from one polarity depending on how you wire it.
400mA is a lot of hotrodding for the DCB1. You can set it up for as little as 50mA if that is what you want.
Pass DIY Addict
Joined 2000
Paid Member
My DCB1 uses a 15-0-15 100VA transformer and is bias at 226mA and I'm running an extra regulator to power my LDR-based volume control. The entire setup draws .110A on the 120V house mains.
So, ballpark it right now at a total of ~13w power consumption. 50VA should be sufficient for what you are describing.
So, ballpark it right now at a total of ~13w power consumption. 50VA should be sufficient for what you are describing.
I would avoid to recommend in a BOM resistors as expensive as almost one MEZ PCB each, since it can do a fine job even on generic parts.
Else, I could accidentally fix in people's minds that only with $$$ parts its worthy to make it. Then exotic step pots ideas surely come along etc.
It does not take $2000 in parts to have a good DCB1 but I have seen such builds here too. Its better starting with fair parts IMHO.
Someone must know its a keeper to his liking first and then he is welcome to pick our minds further in the threads if determined to invest.