Welcome! Need support, you got it. Or share youe ideas and experiences.
Return to K&K Audio / Lundahl Transformers
68.156.164.139
In Reply to: Re: But there is no regulation for the DAC ... posted by Dave Davenport on December 2, 2004 at 17:34:10:
I am sorry I passed out BAD information.I had come to the same conclusion about putting the tube in "backwards", especially since I am not using the rectifier tube on the board. Thanks, though.
![]()
Follow Ups:
Not using the rectifier??Don't tell me that you are going to use sand!!
And lots of capacitance after the rectifier!!
:-))
I am using some J.C. Morrison recommended 6DM4 damper tubes.I must admit my set-up does include a choke input, old habits die hard, but there is the requisite MINIMAL capacitance after the choke. No one can question the value in doing that, especially when using a current source, as both you and Mr. Wright have done, before the voltage regulator.
I was surprised that you are not using a "comparator" within your shunt regulator. Did you try this and decide the complexity was not worth it, or more importantly, it did not sound as good?
ALWAYS CURIOUS!!!
Thanks,
Rick McInnis
Hi Rick,Okay on the choke, just make sure that the input voltage to the CCS is high enough. It can have lots of ripple but the negative peak of the ripple should be at least 20V more positive than the voltage level at the output of the CCS. As an experiment, you might try it without the choke.
There is an interesting story with the shunt regulator. The basic design of the Minimal Reactance power supply does not have a shunt regulator on the output. The design is that the CCS sources a fixed current, say for example 35mA. The load draws a fixed current, say for example 30mA. The difference, or 5mA, flows through a resistor on the output. The size of this resistor sets the B+ voltage, say for example, 40K at 5mA sets 200V. The value of the output voltage isn't important as long as it is high enough to keep the CCSs in the load in regulation.
The adjustable factors in this are two currents and a resistance. Getting the right mix can be tricky: kind of like balancing a plate on a stick. Once everything is set the way you want it is stable and provides the needed regulation.
This is fine if you are building one but is a problem for production. The problem is that each batch of MOSFETs has slightly different characteristics. Therefore for each batch you have to empirically choose a resistor to set the source current, a resistor to set the load current and a load resistor.
So to get around having to do this I added an "automatic variable resistor" to the output in place of the fixed resistor. This is the shunt regulator. Because of its function, to keep the voltage high enough to drive the CCSs in the load, it does not need to have a precision reference.
how you were accounting for the differences in the FETS.INGENIOUS, but then, I am not surprised.
Would you be willing to advise of a source for the FETs you are using OR can I substitute the SUPERTEX dievices I got from K & K?
What can be done to formalize a process for the USB input board? Would Mr. Carter be willing to add a spot on his digital page for those interested? I am afraid after a few more messages are posted the idea will be lost "below".
Merry Christmas,
Hi Rick,For MOSFETs you can use either SUPERTEX DN2540 from K&K or IXYS 10M45 and I get them from All American supply.
As for the USB dac, I am doing a little studying in the background but IF we decide to do a formal project, that will have to wait until after we return form CES the second week in January.
This post is made possible by the generous support of people like you and our sponsors: