DCB1 Build

[omishra]

There is trick to solder this kind of switch.
1. Clean (with blade or sand paper) and apply flux to metal part of switch.
2. Heat it quickly and melt some solder on it. If metal is really clean then solder will stick to it.
3. Strip and apply flux on the clean metal on wire end.
4. Heat and melt solder material on it. If wire metal is really clean (free from grease and oil) then solder will stick to it.
5. Hold both together in position and heat quickly. It takes <2 sec to join them. Hold it till it solidifies.

 

[omishra]

From diya forum. Wrt pot selection.
50K makes narrower bandwidth when mating with the 2SK170 input capacitance than with 20K or 10K. Technically the bandwidth is still wide enough with the 50K but I could listen to a bit more cleaner result with the smaller value pots.

 

[jls001]

20 or 10K linear or log? I have a 20K linear and can give it a try.

 

[omishra]

Log is preferred (thinking power increment requiring for each level) for audio volume control but linear does not matter wrt quality. Only scale of loudness at each level will be different.

 

[keith_correa]

Can't we approximate a log taper with a linear pot? I think I read somewhere that it can be done.

 

[omishra]

Yes, it's at Rod Elliot's website http://sound.westhost.com/pots.htm

It's resistor between Wiper and ground of pot. 15K-22K

 

[sann]

Since we are at volume control, please let me know how to limit the volume to say 50/60% of the current full level, this is to avoid risk of turning the volume pot with high efficiency speakers by kids.

 

[omishra]

Put value equal to resistance of pot, in series of wire at CW pin. But that would be good only for low impedance pots, 10K-25K. Not advisable for pot of 50K.

Btw I had accidentally run my GR research speaker on my Tandberg 3003 power amp with full volume. It's sounded good, did not harm except everyone in house running towards bedroom where I was playing. why not? It's rated 250W/ch on 4ohm and 165W/ch on 8ohm. You imagine how much I did pump up into it. for next 3 hrs there was whistle blowing through my ears.

 

[jls001]

I swapped out the 47k CITEC log pot from the DCB1 with a 10K CITEC log pot. I thought I had a 20K linear, but it turned out to be 10K log. I think element14 messed up my order.

Anyway, I don't like the sound of it.

The soundstage has narrowed than before, though it has become a bit deeper. It sounds bunched up around the center, and instrumental separation has suffered. Also, the bass has lost its mojo. Bass was hard-earned in my setup, and had improved incrementally through lots of incremental changes. So this omission is, in itself, a big enough sin warranting, prima facie, the ejection of the newly intro-ed component. Don't know if pots need burn in. May be I'll play a few days and see if things get better.

I bought a 22K linear pot (some brand called Elcom) but unfortunately the shaft is too short for my volume knob. May be I should look for an appropriate sized knob for it.

 

[jls001]

@Om: a few questions:

1) Do you think using small caps across the RCA input ground and the chassis, as used in some device would be beneficial for RFI/EMI rejection? I have seen this often in pro audio devices and even a phono pre. If it has benefits, I'm perfectly ready to implement it on the (two) inputs and output of my DCB1 (and may be even phono stages).

2) The Audiocrafts stocks 50K log ALPS Blue Velvet pot. Any idea if it will be an improvement on the stock CITEC (UK) 47K pot? Audiocraft's cost is reasonable (Rs 280).

3) For magma's DCB1 build we are planning to have two inputs and use pull down resistors of about 15K in one of the inputs as his DAC's o/p impedance is rather low. Has anyone tried this method to present a constant input impedance? Does it work well in practice? My limited understanding is it is used more in logic circuits, but even in that application it usually precedes a buffer stage.

Thanks in advance.

 

[omishra]

1. Yes, that is standard for RFI/EMI SUPPRESSION. I use it for wide band amplifiers and switched amplifiers like class D/class T. I use 470pF-1KpF for small signal jack. Even I used 4700pF on speaker binding posts to earth ground of class T amp chassis. These all are ended nearest to chassis and entry/exit points. I had read few papers on this. Btw correct grounding is must for this to work. Like earth and zero volt in ps must be handled correctly.

2. You can try, Alps blue. It would get better longitivity, balance and tracking. Signal does not know through which type of resistance it is travelling. but other factors in pot like wiper contact quality, stray capacitance and inductance matters somehow.

3. I would say leave it alone to default. Golden rule - output impedance must be low, ideally 0 and input impedance must be very large, ideally infinite. Your DAC and DCB1 already follow these rules.

 

[jls001]

Funny, another builder reported losing out on bass punch when changing pot from 50K to lower. See post #4653.

http://www.diyaudio.com/forums/analog-line-level/171715-salas-hotrodded-blue-dcb1-build-466.html

 

[sann]

This was the main reason for me to add a buffer before the LDR volume control having 13K impedance.

 

[omishra]

This was the main reason for me to add a buffer before the LDR volume control having 13K impedance.

I am using it same way because LDR volume control is best of all - no Wiper, no mechanism or housing impacts, pure variable resistance, directly varied from voltage controlled light. Only problem comes with effective impedance is 13K which on lower side of sources desired. So let it drive by buffer rather than source. Before LDR I tried Alps, ladder based stepper, "shunted pot-fixed series" volume control etc.

 

[jls001]

1. Yes, that is standard for RFI/EMI SUPPRESSION. I use it for wide band amplifiers and switched amplifiers like class D/class T. I use 470pF-1KpF for small signal jack. Even I used 4700pF on speaker binding posts to earth ground of class T amp chassis. These all are ended nearest to chassis and entry/exit points. I had read few papers on this. Btw correct grounding is must for this to work. Like earth and zero volt in ps must be handled correctly.

@Om: what is the ideal value of capacitor to use? Does voltage rating matter, as we're dealing with line level signal here? I want to try it out on the DCB1 so I will go get some caps from the market. I have a good number of Keltron 100 pF caps right now but I am not sure if that's the best value to use.

For earth on the DBC1 I have not implemented the rectifier-thermistor ground isolator, but I can do that too if you feel it could be beneficial. However, I have ensured that the returns of the RCA in/out sockets are completely isolated from the equipment chassis. Is this isolation sufficient, or should I add the ground isolation?

 

[omishra]

@Om: what is the ideal value of capacitor to use? Does voltage rating matter, as we're dealing with line level signal here? I want to try it out on the DCB1 so I will go get some caps from the market. I have a good number of Keltron 100 pF caps right now but I am not sure if that's the best value to use.

For earth on the DBC1 I have not implemented the rectifier-thermistor ground isolator, but I can do that too if you feel it could be beneficial. However, I have ensured that the returns of the RCA in/out sockets are completely isolated from the equipment chassis. Is this isolation sufficient, or should I add the ground isolation?

See idea here is to ground the airborne noise which may pollute our signal input ground through source ground wire. It is best referred to earth ground. because you are shunting input ground before it comes to PCB ground plain.

I used 25V ceramics at RCA and 100V at speaker posts. But here its okay with ceramics. There is not ballpark value for these caps. It depends upon noise which you want to protect from. e.g. class T has internal clock of 700KHz-1.2MHz, so I need speaker wires not emitting anything of it or its harmonics. That is 4700pF/100V-250V.

DCB1 RCA input (live and ground) could be using 15+100pF on live and 15+1KpF on RCA ground to chassis.

 

[jls001]

DCB1 RCA input (live and ground) could be using 15+100pF on live and 15+1KpF on RCA ground to chassis.

So should I parallel two caps as per above values?

1) 15pF + 100pF for live
2) 15 pF + 1000 pF for gnd

I didn't know that we should use on live lead too!

25V rated caps should be fine?

And not needed for output, right?

 

[jls001]

2. You can try, Alps blue. It would get better longitivity, balance and tracking. Signal does not know through which type of resistance it is travelling. but other factors in pot like wiper contact quality, stray capacitance and inductance matters somehow.

3. I would say leave it alone to default. Golden rule - output impedance must be low, ideally 0 and input impedance must be very large, ideally infinite. Your DAC and DCB1 already follow these rules.

@magmaji, pls see Om's opinions above.

We can use the CiTEC 47K log pot Sachin included in the kit, or you can try the ALPS pot. If you're ordering, do include one for me too we'll settle a/c later.

Point 3 we can still try if you wish. It should be easy enough to dismantle if we don't like it.

 

[hifiramr]

@magmaji, pls see Om's opinions above.

We can use the CiTEC 47K log pot Sachin included in the kit, or you can try the ALPS pot. If you're ordering, do include one for me too we'll settle a/c later.

Point 3 we can still try if you wish. It should be easy enough to dismantle if we don't like it.

You will get ALPS pot from the indian site save shipping amount from outside

theaudiocrafts

 

[jls001]

You will get ALPS pot from the indian site save shipping amount from outside

theaudiocrafts

Thanks, hifiramr. In fact we are planning to buy from audiocrafts.