Tracking preregulator.
Remove LED4, it is fighting against Q1.
With R2 = 100R, you are running 12.5 mA over Q1. Increase R2 to 1K to decrease current to 1.25 mA. R5 and C6 are forming LPF, you can increase R5 to improve filtering. C6 should be film capacitor or low leakage electrolytic.
And don't forget to load down reg output with 56 - 100R 1W (or more W) resistor. It will give you lower output impedance and you get free shuntie parallel to your load.
If LED4 is removed, then your output voltage will be ~ 3.7V Increase R8 to 13K to dial output voltage to ~ 3.3V
http://www.diyaudio.com/forums/power-supplies/28978-improving-lm3x7-regulator-circuit-4.html#post356154
Using 3-pin regulators off-piste: part 4
Remove LED4, it is fighting against Q1.
With R2 = 100R, you are running 12.5 mA over Q1. Increase R2 to 1K to decrease current to 1.25 mA. R5 and C6 are forming LPF, you can increase R5 to improve filtering. C6 should be film capacitor or low leakage electrolytic.
And don't forget to load down reg output with 56 - 100R 1W (or more W) resistor. It will give you lower output impedance and you get free shuntie parallel to your load.
If LED4 is removed, then your output voltage will be ~ 3.7V Increase R8 to 13K to dial output voltage to ~ 3.3V
http://www.diyaudio.com/forums/power-supplies/28978-improving-lm3x7-regulator-circuit-4.html#post356154
Using 3-pin regulators off-piste: part 4
Or use a 7805 followed by a ADP150/151 + necessary caps and experience a very low noise 3.3 V supply (max. 150/200 mA). Just 9 uV noise, less parts, less hassle. Given the fact that a BC547 is used you will not exceed 100 mA of current anyway so ADP150/151 should be OK.
Are those real measured values or calculated ones ? I can't imagine picoVolts of noise with LM117.
Are those real measured values or calculated ones ? I can't imagine picoVolts of noise with LM117.
Last edited:
Thanks for the suggestions.
The bc547 transistor is limited to 100 mA, is there a popular substitute that could provide 200 mA?
In the circuit with transistor output that reduces the noise, different load greatly increased noise.
Is not it better to use the second circuit with the leds?
The bc547 transistor is limited to 100 mA, is there a popular substitute that could provide 200 mA?
In the circuit with transistor output that reduces the noise, different load greatly increased noise.
Is not it better to use the second circuit with the leds?
Attachments
OK, the schematic makes not much sense to me. Why use a noisy reg that can handle 1.5 A preceded by a BC547 that can handle 100 mA ?
You can build the circuit but don't be surprised that for instance a plain Jane 3.3 V reg like LF33AB has better specs. Some of the newer SMD LDO 3.3 V regs like the ones mentioned before are even way better.
You can build the circuit but don't be surprised that for instance a plain Jane 3.3 V reg like LF33AB has better specs. Some of the newer SMD LDO 3.3 V regs like the ones mentioned before are even way better.
Last edited:
Aha, there is an explanation how standard 317 or 78xx 3 pin regs can perform better. Also the PFM Flea is mentioned which is an excellent reg. I use those on a regular basis and they perform very good, also on 3.3 V but max. current is quite low (enough for a lot of modern DAC chips but not for the older bipolar chips).
You can choose to improve a standard LM317 circuit but choosing another reg may be simpler.
You can choose to improve a standard LM317 circuit but choosing another reg may be simpler.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.