VSSA Circlotron

Status
Not open for further replies.
Jump to Latest
#61
#62
or like a kaneda

kanedaa.gif
 
#68
UltimateX86 said:
Hi,

I think there are too many capacitors

If you want a simple schematic with only two stage you could made a mode of this with 2SK1058 : http://www.diyaudio.com/forums/solid-state/270239-i-call-infinitron.html

I will try to sim
Click to expand...

Hi UltimateX86 !

Thank you for your response! Yes you are right, too many capacitors in the signal path. The Infinitron circuit looks nice and quite simple. I checked the LME49830 datasheet and apparently is an excellent device for building neat and simple amplifiers. I did a search on the Net and found a lot of amplifiers modules using this remarkable chip. They are Chinese of course. What do you think of these Chinese amplifiers ?
 
#69
Joe Berry said:
It's basically the circuit UltimateX86 first presented, but with differential feedback applied as described in post #5. There is one other difference, which is that the output stage uses two HEXFETs biased as shown in the Thorens TEM3200 conceptual schematic that you see out on the web.

I first saw this bias arrangement in a BJT circlotron design that appeared in No.8 1984 of RadioHobby, a Russian magazine. It was essentially a lower-power version of the Sumo Nine, with the important difference that the bias current is regulated by a constant current source rather than resistors. This allows for the possibility of temperature compensation, and with that, for the use of either BJTs or high-gm MOSFETs as output devices.

However, this arrangement is not so good for use with lateral MOSFETs in the output, because it limits their drive voltage to twice the bias voltage, which is not nearly enough for full output current. It would probably work well with power JFETs, though.
Click to expand...

Hello,

No.8 1984 of RadioHobby, a Russian magazine

where can a found this article ?

thanks
 
#74
It is looking good, including the feedback arrangement. All JFETs benefit from cascoding. 8 mA current will be far too much.
The references introduced greatly support operational stability. In addition, unbalanced input would be a better option, reducing the value of R8 and R21.
R3 is not needed. I would increase the value of R2 and R5 to something like 100 ohm. The compensation capacitors will probably be unnecessary. What about a carbon resistor on the source electrode of the MOSFETs?
 
#76
N101N said:
It is looking good, including the feedback arrangement. All JFETs benefit from cascoding. 8 mA current will be far too much.
The references introduced greatly support operational stability. In addition, unbalanced input would be a better option, reducing the value of R8 and R21.
R3 is not needed. I would increase the value of R2 and R5 to something like 100 ohm. The compensation capacitors will probably be unnecessary. What about a carbon resistor on the source electrode of the MOSFETs?
Click to expand...

8mA is the Idss, I use 5.3mA.

what voltage to apply to the cascode?

unbal input ? I have balanced DAC/preamp and LL1690 especially for the input circlotron 🙁

why increase R2 and R5 ? thermal stability ?

Usually the lateral MOSFET do not need it, especially that there will be no paralleling
 
Last edited:
#77
X

just for input JFET pair is not the problem since in that configuration their drains are 100% bootstraped , but than care must be taken that max. Vds to be not exceded ,
on the other hand both VAS stage virtually stays beetwen generated AC of balanced output lines , with colectors 100% boostraped(OK), but in the same time with emiters which floats on 100% inverted AC lines(bad).
 
#79
separate B+ supply line just for the VAS stages , but that separate B+ line can be derived from OPS B+/B+`lines, via two resistors and one smoothing electrolytic capacitor connected to ground ,it is somethig like this :
 

Attachments

  • circlotron1.4A (1).png
    circlotron1.4A (1).png
    35.2 KB · Views: 428
#80
why increase R2 and R5 ? thermal stability ?

Usually the lateral MOSFET do not need it, especially that there will be no paralleling
Click to expand...

To stabilize the base-emitter and gate-source junction, lowering distortion.

unbal input ? I have balanced DAC/preamp and LL1690 especially for the input circlotron
Click to expand...

I see. OK.

5.3mA is still too much to me.
Could you implement the cascodes?
 
Status
Not open for further replies.
Top Bottom