My take on a discrete shunt voltage regulator

[HiFiNutNut]

The set up is as follows:

v2 published a month ago (not the latest), built as per the schematic using perf board. Made shortest possible connections between transistors.

Tx (15V-0-15V) - 1N5822 - 3 x Panasonic FC 3,300uF || 1 x Rubycon ZL 2,200uF

1.1uF, 20k, 20k Vref

CCS current set to just above 100mA

Remote sensing.

22uF/50 Rubycon ZL (Imp=0.34R at 100kHz) at the output.

[HiFiNutNut]

Replaced the LM4562 with the OPA627. That high frequency uneasiness I reported last night was gone.

In terms of regulators, this is the best I have had so far. The oscilloscope I got on Monday helped.

I will spend the next 2 days to select the best output capacitor (and its ESR) and then wrap this project up. I will then use this preamp to (re-)do speaker driver measurements, and will within weeks complete my active speakers.

Thanks to everyone who contributed, especially Ikoflexer and Salas.

[HiFiNutNut]

My oscilloscope can only display 10mV a grid and 0.5uSec.

The thin flat line I referred to must be under 1mV. It did not look any thicker, actually a bit thinner than the screen picture from Ikoflexer in post 196, but with 10mV per grid. I wish the resolution of the scope can be higher.

One thing I noticed was that the line obviously brightened up when it was set to 5uSec, which corresponds to 2MHz per grid. So there may still be resonance at this region, but it must be very low in amplitude, no more than or at most 1mV for sure.

[HiFiNutNut]

Ikoflexer,

I know you have inserted nH into capacitors and other tracks. Have you inserted some nH between Q1 and M1 via R4? This is where very low impedance output going into very high impedance input, including R4, the wire length must be at least 1.5cm so that is a 10.53nH at a minimum! I have not soldered R4 on the upper leg of M1, so I expect the inductance may be as high as 20nH! Could you please have a check?

Regards,
Bill

[HiFiNutNut]

Originally Posted by ikoflexer

tested a 4uF film cap on the output and you can see the attached resulting trace images.

4uF? Odd value. You mean 2uF || 2uF? or a 3.9uF? What capacitors are they?

[Melon Head]

Please forgive me for what I am about to say.

Has anyone tried feeding the Salas reg with a switch mode power supply?

[ikoflexer]

Originally Posted by HiFiNutNut

When the scope probe was at the rails, whether at the regulator output end or at the load end, the screen displayed a thin, flat, straight, static line.

I thought I had to be careful so I checked very carefully and did it many times. The result was confirmed.

[ikoflexer]

Originally Posted by HiFiNutNut

One thing I noticed was that the line obviously brightened up when it was set to 5uSec, which corresponds to 2MHz per grid. So there may still be resonance at this region, but it must be very low in amplitude, no more than or at most 1mV for sure.

Just an artifact of the scope. I think there would be more signs if you had oscillation.

[ikoflexer]

Originally Posted by HiFiNutNut

4uF? Odd value. You mean 2uF || 2uF? or a 3.9uF? What capacitors are they?

Got it from a surplus store here in TO. Says "Ero MKC 4uF 200V" on it. Ugly green colour :d

[ikoflexer]

Originally Posted by HiFiNutNut

Ikoflexer,

I know you have inserted nH into capacitors and other tracks. Have you inserted some nH between Q1 and M1 via R4? This is where very low impedance output going into very high impedance input, including R4, the wire length must be at least 1.5cm so that is a 10.53nH at a minimum! I have not soldered R4 on the upper leg of M1, so I expect the inductance may be as high as 20nH! Could you please have a check?
Regards,
Bill

Checked with a 25nH, doesn't change a thing.