Salas hotrodded blue DCB1 build

Jump to Latest
O

Offie

Member
Joined 2014
jeg elsker deg :)

friends from norway:) great to hear that you're B1 is also working nice.
As attanuator I use a 10k. If you're sources can handle this....try it!
An externally hosted image should be here but it was not working when we last tested it.


But 25k is more ..lets say the save way to load your sources. But I think lots of the sources nowadays can easely be driven by a 10k load.

I'm even gonna try a factor 1/4 of impedance matching...and see what will happen. the old style "le pacific" phono pre has a high output impedance. Modyfying the circuit (and lowering its gain will help me)
 
pellesmil

pellesmil

Member
Joined 2010
Thanx Offie.

Nice clean build you have there.
Whats the bunch of wires on the RH side ?
I guess you are not running more than one input so its not input selection wires?

I purchased my auttonator from Hificolective a couple of ears ago discussing ladder vs Shunt type, after some advice from Nick i went for a ladder type.
The impedance though at 10K was based on my main source at the time :
Analogue Addicts Phono Preamplifier 2006 Edition.
Now adays i am using Salas folded that should have no problem driving 10K.

At the moment i run my setup passive with only the auttonator as a volume control feeding a Holton hpa-nxv 201ps with 33Kohm input impedance.
I am expecting an upgrade going to the DCB1 whit its low output impedance.
 
A

AndrewT

R.I.P.
Joined 2004
yoshi788 said:
Thank you Salas.

What's the output impedance of the DCB1 in balanced mode (one DCB1 per channel)? I'm trying to find technical explanations to the different sound with and without the DCB1 in the set up. My Lynx card is 100 ohms in balanced mode btw.
Click to expand...
add the two output impedances together to get the total output impedance of the dual version.

Balanced impedance connections give better interference rejection, if the ratio of load impedance to source impedance is MUCH greater than the usual 5:1 to 20:1
W.Jung suggests 50r:~1M, i.e. 20000:1
 
Salas

Salas

diyAudio Chief Moderator
Joined 2002
Paid Member
yoshi788 said:
Thank you Salas.

What's the output impedance of the DCB1 in balanced mode (one DCB1 per channel)? I'm trying to find technical explanations to the different sound with and without the DCB1 in the set up. My Lynx card is 100 ohms in balanced mode btw.
Click to expand...

As Andrew said, balanced Zout is always double the single ended figure. So you got about 500R now including the JFETs internal. Use the smallest value for output resistors that is not allowing oscillation in your system for the sound to become a bit bolder.
 
Y

yoshi788

Member
Joined 2014
Salas said:
As Andrew said, balanced Zout is always double the single ended figure. So you got about 500R now including the JFETs internal. Use the smallest value for output resistors that is not allowing oscillation in your system for the sound to become a bit bolder.
Click to expand...

Thank you AndrewT and Salas. I think that explains much of the difference with and without the DCB1 in my setup: The Lynx card's output impedance is ~ 5 times lower. I guess that the output impedance should be 100 ohm or less in my case, to be comparable to the Lynx. Do you think replacing the 220R resistors to 50R is a sensible starting point?
My DCB1's are driving a FW F1J clone. According to F1J specs, input impedance is 10kohm.
 
Salas

Salas

diyAudio Chief Moderator
Joined 2002
Paid Member
I have seen stable ones with 50R or 100R. Has to do with interconnects capacitance too. If having a scope you may hook up the interconnects to dcb1, terminate with 10K at their ends, and test for squarewaves without fuzz there. If without such gear, in the real system the oscillation may break out like an HF squeal. Sometimes low in volume, or it can be quite high depending on the amp, so verify with cheap test speakers like plastic ones from some old home PC, to not fry any main system tweeters.
 
O

Offie

Member
Joined 2014
Hi pellesmil
Thos bunch of wires on the right hand side are indeed for input selection. On this moment I don't use it...but maybe in the futere. I'm also gonna add a "le pacific" Jfet phono in here....hope there is room enough...and sheilding enough.
I'm really happy with the DCB1! Salas and all the rest did a fantastic job!
Even the rest of matched pairs I can get rid of. But I would like to keep some for myself for future projects. The 2SK170 is really fantastic!
I think you're idea of a ladder type attenuator is better than the series I use. Mine is from an old Burmester preamp.
 
Y

yoshi788

Member
Joined 2014
Salas said:
I have seen stable ones with 50R or 100R. Has to do with interconnects capacitance too. If having a scope you may hook up the interconnects to dcb1, terminate with 10K at their ends, and test for squarewaves without fuzz there. If without such gear, in the real system the oscillation may break out like an HF squeal. Sometimes low in volume, or it can be quite high depending on the amp, so verify with cheap test speakers like plastic ones from some old home PC, to not fry any main system tweeters.
Click to expand...

Unfortunately, I don't have a scope available. Testing in actual system seems quite risky, so I will have to find some test speakers somewhere. Or maybe this is a good opportunity to buy a scope and a signal generator:cool:
 
M

mbain

Member
Joined 2011
AndrewT said:
Isolate the RCA from the Chassis.
Place the Left and Right RCA sockets very close together.
Use twisted pair to connect RCA to PCB.
At each end of the twisted pair ensure by careful placement of the wires that you have a very small loop area for least pick up of interference.

If you decide to "Ground" the input at the PCB then run the ground wire/trace to the Main Audio Ground (MAG) over the route that the other Flow Current follows through the PCB.

If you "Ground" the input at the RCA, then run the ground wire along the twisted pair to the PCB and then follow the Flow route of the other PCB currents to the MAG.
This RCA located "Ground" wire can be two wires, one from each RCA or you can couple the two RCAs and run one Ground wire back to MAG.

The input wiring is not "simple", it needs to be thought out and run carefully to maximise it's ability to reject interference. Adopting a coax does not make it any easier. In some respects a coax is more difficult.

When you come to wiring the three PSU wires, it requires just as much thought for correct routing.
Similarly when you wire up the Speaker Terminals.
All these connections to the "amplifier" are potential INPUTS and OUTPUTS and ALL need careful routing.
Click to expand...
Andrew,
I would appreciate a reference or a more basic/simplistic explanation of your earlier response to my question re. wiring. To clarify I am building a one input, Goldpoint attenuated DCB1 buffer. I become confused from line four of your response until the end. Please help!
 
M

mbain

Member
Joined 2011
Thanks for the help. I want to do this as correctly as I can. Beginning with the input RCA plugs; I should take a wire from the right and left center pins, twist them and connect to the attenuators right and left inputs. Correct? Now, what to do with the RCA right and left input ground tabs? Then, from the right and left attenuators output tabs, connect a wire to each tab, twist and connect to the PCB,s molex right and left input. Correct? From the PCB molex output, take a wire from the right and left pins, twist and connect to the center pin on the right and left RCA output plug. Correct? Remaining questions. What to do with the right and left ground tabs from the attenuator? I suspect I should connect one wire to each of the right and left ground tabs, twist and connected to the Molex PCB input center pin. Correct? What to do with the ground pin from the Molex output center pin? Connect two wires, twist and connect one to each ground tab on the ouput RCA plugs? This is enough for now. When it comes time to wire the transformer I would like your help again. Thanks for your assistance.
 
A

AndrewT

R.I.P.
Joined 2004
mbain said:
............ Beginning with the input RCA plugs; I should take a wire from the right and left center pins, twist them and connect to the attenuators right and left inputs. Correct?
Click to expand...
No.
taking only the left RCA. make up a twisted pair, one core for the Hot from the Hot pin of the RCA socket, one for the Return from the outside of the RCA socket. Take this twisted pair to the next stage. Here the Hot connects to In and Cold/Return connects to Signal Return. NOT to Power Ground and NOT to MAG. At each end of this twisted pair, ensure that the LOOP formed by the two wire ends is as small as possible. This requires the Hot and Cold/Return to be very close together, both on the back of the socket and at the input to the next stage. Repeat all this for the right channel.
Now, what to do with the RCA right and left input ground tabs?
Click to expand...
the twisted pair are a Flow and Return of the signal. There is no ground wire ! and no ground tab !
Then, from the right and left attenuators output tabs, connect a wire to each tab, twist and connect to the PCB,s molex right and left input. Correct?
Click to expand...
identify the two input terminals of the vol pot. connect the Hot and Return for the left channel. Connect the Hot and Return for the right channel.
From the PCB molex output, take a wire from the right and left pins, twist and connect to the center pin on the right and left RCA output plug. Correct? Remaining questions. What to do with the right and left ground tabs from the attenuator? I suspect I should connect one wire to each of the right and left ground tabs, twist and connected to the Molex PCB input center pin. Correct? What to do with the ground pin from the Molex output center pin? Connect two wires, twist and connect one to each ground tab on the ouput RCA plugs? This is enough for now. When it comes time to wire the transformer I would like your help again. Thanks for your assistance.
Click to expand...
This next para seems to be confused because of the earlier misunderstanding of "ground" tabs and "ground" wires. There are no "ground tabs" in the signal wiring. You only have Flow and Return. These must ALWAYS be run as close coupled pairs.
 
M

mbain

Member
Joined 2011
AndrewT said:
No.
taking only the left RCA. make up a twisted pair, one core for the Hot from the Hot pin of the RCA socket, one for the Return from the outside of the RCA socket. Take this twisted pair to the next stage. Here the Hot connects to In and Cold/Return connects to Signal Return. NOT to Power Ground and NOT to MAG. At each end of this twisted pair, ensure that the LOOP formed by the two wire ends is as small as possible. This requires the Hot and Cold/Return to be very close together, both on the back of the socket and at the input to the next stage. Repeat all this for the right channel.the twisted pair are a Flow and Return of the signal. There is no ground wire ! and no ground tab !identify the two input terminals of the vol pot. connect the Hot and Return for the left channel. Connect the Hot and Return for the right channel.This next para seems to be confused because of the earlier misunderstanding of "ground" tabs and "ground" wires. There are no "ground tabs" in the signal wiring. You only have Flow and Return. These must ALWAYS be run as close coupled pairs.[/QUOTE
Andrew,
This will result in the "R" input at the pcb molex connector having one wire attached(from the R hot rca pin) the L input at the pcb molex connector having one wire attached(from the L rca hot center pin) and the center pin of the pcb input with the molex connector having two wires attached(one for each return to the outside of the rca socket. Yes?
Still don't understand the attenuator wiring as it has a "in", "OUT", and "GND" terminals.
Click to expand...
 
A

AndrewT

R.I.P.
Joined 2004
yes, when the two sets of twisted pairs arrive at the 3pin INPUT of the DCB1 you have to connect the two Return Wires together to feed into the commoned Return pin.
This Return pin is probably labeled wrongly.
When will they ever learn?
The "ground" word causes confusion and not just among beginners.
 
You must log in or register to reply here.
Top Bottom