The simplistic Salas low voltage shunt regulator

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
What is the Vac output range of the transformer on zero load (open circuit)?

Use the lowest voltage from that range to determine the input voltage to the regulator.

I think that the 13Vdrop across the 65r is too high for a 200mAdc current output.
Try adjusting your input voltage, by adding parallel resistors to your 68r (E6 value)

Could you post a link to that spreadsheet?

On the spreadsheet:
The output voltage is specified as 21Vdc. The current as 200mA, the resistor as 65r.
Working back, the voltage on C1 & C2 must be 21 +[0.2*65] = 34Vdc.
That just about agrees with the input voltage.

In PSUDII:
Add in the first capacitor.
At present the 65r is preventing the smoothing capacitor from charging up. It's working a bit like a choke input filter without the choke. The resistor is "choking" the output.
 
Last edited:
Member
Joined 2006
Paid Member

Attachments

  • PSUII.png
    PSUII.png
    85.5 KB · Views: 634
Last edited:
The RC time constant tells you how fast the cap charges up.
One RC gets just over 60% of charge. Five RC gets >95%. Ten RC is effectively fully charged.
RC = 120mF * 65R = 8seconds.
Check the plot. is the voltage on C2 >60% of final voltage at 8s?

Check the plot at 40s, Is the voltage >90% of?

Check the plot at 80s.

New RC = 65 * 470uF
10*RC ~ 13s for virtually final voltage.

But, you have to correct all these RC for the effective resistance of that 200mA load.
Just accept the PSUDII predictions, it's easier that way, provided you ask the simulator the correct questions.
It comes back to the never too old advice: Never use, nor rely on simulation software, unless you can check that the answers are going to be correct. The simulation is intended to save time. It does not teach electronics basics.
 
I have stock of BC550 and BC560 for Salas shunt regulator

Sellers said that these are real Philips. Can anyone tell me is these real or fake?
An externally hosted image should be here but it was not working when we last tested it.

I have these too
An externally hosted image should be here but it was not working when we last tested it.

The Batch number start with "BK". Any information on the manufacturer?

Best Regards
 
I can't figure this one out:

I built a 1.1 (same schematic as BiB, just on my own pcb). Set it to Vout=5V (unloaded).

When I add a 100R load, the Vout sags to under 1V. The CCS current is 200mA, the Vin=14V.

I am using IRF9610 and MJE15030. What's going on here? The CCS seems to be working correctly, so why still the voltage drop under 100R load@5V ?
 
diyAudio Chief Moderator
Joined 2002
Paid Member
2sk170 and bc547

Hmm... Scope it if its surely correct for build. Don't know about K170 in a V1.1. I hope you don't get any oscillation because of different (higher) fet capacitance or higher IDSS on the error amp bias. The pcb BIB with K117 has been used so many times at 5V both with BJT and Mosfet shunt, and there is no bad reaction I am aware of.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
The f+ and s+ connect at the load. The f- and s- connect to a ground plane ( the load, tda1541a, is on the same pcb as the reg)

This can be a potential problem. Kelvin is for defined node. Do an experiment. Short F+ & S+, also F0 & S0 locally at the reg's output nodes. I.e. make it a 2 wire reg. Then try 100R on those two wires from those nodes. See if it behaves better.
 
This can be a potential problem. Kelvin is for defined node. Do an experiment. Short F+ & S+, also F0 & S0 locally at the reg's output nodes. I.e. make it a 2 wire reg. Then try 100R on those two wires from those nodes. See if it behaves better.

This is in principle how my pcb is layed out: f+ and s+ connect near the load (which is 1cm from the reg, really). F0 and s0 both connect to the same gnd plane, so will follow the path of least restance/inductance to each other. (probably 2cm in my layout)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.