The simplistic Salas low voltage shunt regulator

Status
Not open for further replies.
Jump to Latest
#4,523
P.S. Won't go lower Vo than Vgs+2Vbe+Led Vf mind you.

Edit: This arrangement seems OK for Vo=5V, and that's it. No lower no Led.
 

Attachments

  • Reflekt5Vo.gif
    Reflekt5Vo.gif
    4.6 KB · Views: 629
#4,526
vzs said:
Thanks for pointing this out. I was also calculating that with Led I can't get lower then ~6.5V.

Btw... I hope practically I could get it ~50 nV/SqrrtHz
Click to expand...

Yes, the models are not very accomplished for such predictions. On the other hand I have seen people juxtaposing LTSPICE noise analysis with results in phono preamps claiming a good match. Not sure, I lack lab analyzers to confirm. Especially VLF 1/f values.
 
#4,527
Hi Salas

I am planning to build a bench psu that should be able to work from 10v to 40Vout.

The schematic should be the one below.

I am planning to add a resistor circuit (instead of the trimmer) as per Andrew´s sugestion. How low should this Trimmer equivalent resistance be to get 40Vout ? ( I presume that the higher the resistance, the lower the Vout)
 

Attachments

  • 1V2RposMedium.gif
    1V2RposMedium.gif
    9.4 KB · Views: 608
Last edited:
#4,529
RCruz said:
Hi Salas

I am planning to build a bench psu that should be able to work from 10v to 40Vout.

The schematic should be the one below.

I am planning to add a resistor circuit (instead of the trimmer) as per Andrew´s sugestion. How low should this Trimmer equivalent resistance be to get 40Vout ? ( I presume that the higher the resistance, the lower the Vout)
Click to expand...

Depends on IDSS of each K170. You got to test a range of resistors for given IDSS. But I would not use a CCSed shunt for bench supply. Because you would need to change CCS current ceiling also to be really flexible, and there will be heat when covering the low region when the trafo should cover the high region too. Due to some extra current allowance.
 
#4,530
I will use a TX with several taps. It can output 20V to get 10Vout, 30V to get 20Vout, 40V to get 30Vout and 50V to get 40vout.

I will use 7,4mA Idss for the Vref CCS jfet.

That is not an issue. What I need to know is the lowest Trimmer value to get 40 Vout.

I know that output voltage = VbeQ7 + [IdQ9 * R11] + [IbQ7 * R10] but I do not know VbeQ7, IdQ9 and IbQ7.

In this case we control Vout by modifying IdQ9 that lowers with increased Trimmer value. That is ok if I use only one 1K trimmer, but if I want to use Andrews resistor layout (including a series resistor to the trimmer) I must know what are the limits.

To get lower Vout, there is no problem, but to increase Vout, there must be a ceiling for the lowest trimmer value.
 
Last edited:
#4,531
Create a 0.6V source (maybe a battery, 100R, and a diode), connect your particular K170 across the diode. Use 1k trimmer between GS connected as in the schematic you posted. Note the value that will give you 0.83mA. Use that as your smallest resistor for a ~40V ceiling when in the actual circuit.
 
#4,534
I opted to use one single 1k trimmer as per your design in order to do not loose fine adjustment possibility.

Now I have one issue: I am using R11 = 30k1 and the lowest possible voltage I get is 10.8 V. I get this voltage with the trimmer set to 0 ohm so I can not go lower.

I do not have anymore trimmers... can I get R11 to 27k so I can adjust to 10Vout, without compromising performance ?
 
#4,536
Correction:

RCruz said:
Now I have one issue: I am using R11 = 30k1 and the lowest possible voltage I get is 10.8 V. I get this voltage with the trimmer set to 1000 ohm so I can not go lower.
Click to expand...

The trimmer is 1k so I can not reduce voltage further.

I plan to use 10u MKP.. with 27kr for Vref ... What can I expect if I do not increase this cap value ? More noise than with 30kr? I reduced this R11 from 47k to 30k so I could fine tune the Vout with more precision.
 
Last edited:
Status
Not open for further replies.
Top Bottom