The simplistic Salas low voltage shunt regulator - Page 320

AndrewT · 11th November 2010, 12:12 PM

post3176
What is Vce of Q9?
With high mains voltage could Vce rise above specified rating?

Salas · 11th November 2010, 12:47 PM

He substituted for BC546 he wrote. With 70Vin 50Vout and his current given his voltage over R15 1k in the source he must be having around 40V across Q9 and 50mW I would estimate.

AndrewT · 11th November 2010, 12:51 PM

the problem is that the resistor ladder fixes the emitter of Q9 @ ~24V.
The supply rising to 80Vdc, when mains is high, puts a high Vce across Q9.
I do not think he realises that.
Changing the ratios of the resistor ladder could help alleviate that problem, or simply looking at the range of worst case supply voltages that one expects the circuit to survive with.

Salas · 11th November 2010, 12:55 PM

By stating he put BC546 there (Vceo=65V), he must have had it in mind?

Quote:

Originally Posted by klewis

...and a few modifications for the higher Vin - 6.8k R14 and R4, changed Q9 and Q2 to BC546C from BC550C,...

Ken

klewis · 11th November 2010, 01:18 PM

Quote:

Originally Posted by Salas

By stating he put BC546 there (Vceo=65V), he must have had it in mind?

Hi Salas and Andrew,

Thanks for your help and thoughts. I turned the Vout up to 60, it took a while ( 10 minutes or so) for the Vout to stop climbing up. I adjusted it back down to 60 several times. Measured temp at Q1 at about 50 deg C and the second mosfet at almost 60 deg C. My heat sink was definately warm and is probably too small.

The LED glowed normally. I will hook the shunt up to the scope tonight and see if its oscillating.

Yes, I was concerned that I would exceed the Vceo of Q9 hence the BC546.

Ken

AndrewT · 11th November 2010, 01:25 PM

the cascode (q9) will remove much of the temperature variation in the CCS due to mains voltage changes.
The drift you have found must then be in the Shunt stage or the series FET.
Big sinks will have a profound effect on reducing temperature drift with mains variation.
Further "warm up" drift will tend to zero since the CCS passes by definition constant current.

merlin el mago · 11th November 2010, 03:32 PM

Quote:

Originally Posted by RCruz

Yes.... Those white sonicaps are really excelent.... Similar tone to Mundorf MKP but without the trebble grit.

Platinum teflons maximum value only to 1uF but 0.047uF for RIAA must be excellents

.

klewis · 11th November 2010, 06:54 PM

I went back into the simulation and looked at some of the heat dissipation issues and realized that I have been underestimating just how hot this thing will get. So, several of my concerns were unfounded. For example at this voltage with 100ma load and 200ma current, the shunt mosfet makes over 14w of heat. And my 1k load resistor makes 3.6w of heat. I'm going to take it back to 130ma of current by changing the 3.3 ohm to 5 ohm and see if that doesn't solve some of voltage rise issues.

Ken

RCruz · 11th November 2010, 07:25 PM

Quote:

Originally Posted by merlin el mago

Platinum teflons maximum value only to 1uF but 0.047uF for RIAA must be excellents

.

Might be good but we must consider the size... and the cost.

Salas · 11th November 2010, 08:26 PM

Phew, lots of drawing.

11th November 2010, 12:12 PM	#3191
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	post3176 What is Vce of Q9? With high mains voltage could Vce rise above specified rating? __________________ regards Andrew T.

11th November 2010, 12:51 PM	#3193
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	the problem is that the resistor ladder fixes the emitter of Q9 @ ~24V. The supply rising to 80Vdc, when mains is high, puts a high Vce across Q9. I do not think he realises that. Changing the ratios of the resistor ladder could help alleviate that problem, or simply looking at the range of worst case supply voltages that one expects the circuit to survive with. __________________ regards Andrew T.

11th November 2010, 01:25 PM	#3196
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	the cascode (q9) will remove much of the temperature variation in the CCS due to mains voltage changes. The drift you have found must then be in the Shunt stage or the series FET. Big sinks will have a profound effect on reducing temperature drift with mains variation. Further "warm up" drift will tend to zero since the CCS passes by definition constant current. __________________ regards Andrew T.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

11th November 2010, 12:47 PM	#3192
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	He substituted for BC546 he wrote. With 70Vin 50Vout and his current given his voltage over R15 1k in the source he must be having around 40V across Q9 and 50mW I would estimate.

11th November 2010, 06:54 PM	#3198
klewis diyAudio Member Join Date: Sep 2007 Blog Entries: 36	I went back into the simulation and looked at some of the heat dissipation issues and realized that I have been underestimating just how hot this thing will get. So, several of my concerns were unfounded. For example at this voltage with 100ma load and 200ma current, the shunt mosfet makes over 14w of heat. And my 1k load resistor makes 3.6w of heat. I'm going to take it back to 130ma of current by changing the 3.3 ohm to 5 ohm and see if that doesn't solve some of voltage rise issues. Ken

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)