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.
ikoflexer said:
I congratulate Tham for having the hottest most inefficient class-d t-amp in the world :rofl: :smash:

Reminds me of the saying that "you give with one hand, and take with the other" :D

But worry not, I already have an idea for a simple pic controller based automatic shunt current adjustment that will be more efficient. Would anybody be interested in something like that? Very simplistic :D

The reality is that we don't need to pass so much current most of the time, since we're not listening so loud continuously.

Woaa, that's a real kicker - I'm interested if you're serious
:)
 
ikoflexer said:


Yeah, I'll take a serious look at it. Got myself a PIC chip already, and built a small interface to program it (flash it) from the PC over the serial port. Just ideas so far, no guarantees it'll work.


Understood, but major kudos for the initiative.
I'm playing with Arduino's for preamp control - much easier to get to grips with than PICs, I'm told. Would this be powerful enough here - Atmel 8 bit AVR uC?
 

iko

Ex-Moderator
Joined 2008
No, don't... you wouldn't believe how easy it is. If you want I'll put a page up with a tiny tutorial. Literally, all you need is a serial port on the computer, a few resistors, a couple of diodes, and a capacitor to be able to flash a given program to the chip. To actually run the chip you need one crystal and a +5V source :D

salas, do you think this is way off topic and we should take it to another thread? I wouldn't want to confuse people more than I usually do :xeye:
 
Iko,
Why not start another thread - I'm sure it will be of general interest also - MCUs are becoming more necessary for all of this audio game, DACs, preamps, amps, filters, Vregs, ..........

It would be a good thread to knock around some ideas (only a few - not a smorgasboard of functions) & see how it's programmed & implemented in hardware!
 
Administrator
Joined 2004
Paid Member
Hi ikoflexer,
Please provide a link for a new thread. I think that this is a super idea and a great opportunity for us older hardware smokers. For me, logic is hard wired in SSI and MSI parts. This has got to be easier!

I have a programmer (plus ZIF socket PCB- Sure Electronics) and an in-circuit programmer for PIC chips. I need some help getting started in this. I have an assortment of PIC micro-controllers as well.

So, I have book, PDF files and stuff. My time gets fractured too often, so I need a little help. Once I'm rolling, I'll probably teach myself okay.

I do have the stuff talking to my 'puter.

I chose the PIC platform because many people seemed to be working with it and there were resources. To me, the separate data and instruction pipes make sense too. So yes, I figured it would be easier.

Can you imagine being able to access the comm link in consumer electronics? Program a similar controller and add features, or change them to something that makes more sense. Even if you only use it to make the user interface to a project easier, or cooler, why not?

-Chris
 
Tham, I'm also interested to hear about this as I too want to power a Tripath TA2020 amp and test out both the Salas shunt reg & Ikoflexer's microprocesser control circuit on it (when ready)

I'll be putting it up against some LiFePo4 high current delivery batteries which I have yet to hear on a power amp.
 
Salas, these 3.3V batteries are high current delivery but small (60 Amp continuous & 120 Amp 10 sec burst), low impedance 4-8 mOhm & retain their power delivery right up to full discharge. Heavily used by the remote control aficionados for flying machines due to their weight to power ratio & their lifetime. The only unknowns are noise & how they sound powering amps (I got them for powering DACs, etc.)

Have you tried these?
 
Last edited:
diyAudio Chief Moderator
Joined 2002
Paid Member
Salas, these 3.3V batteries are high current delivery but small (60 Amp continuous & 120 Amp 10 sec burst), low impedance 4-8 mOhm & retain their power delivery right up to full discharge. Heavily used by the remote control aficionados for flying machines due to their weight to power ratio & their lifetime. The only unknowns are noise & how they sound powering amps (I got them for powering DACs, etc.)

Have you tried these?

They seem special. If it beats them it will be no mean feat. Will not up their impedance if you need to chain 4 of them? Build Iko's V1.5 evolution of the shunt since it worked for Tham without weird interactions as V1 did also. It has a performance edge.

http://www.diyaudio.com/forums/showpost.php?p=1890375&postcount=857
 
Ikoflexer, I know that these shunt regs simulate < 1 mohm but how does this translate in the real world - have they been measured at the output of the reg?

I know this isn't the complete quality measure of a power source & I don't know, for instance how the batteries perform up through the frequencies. It might be like in the LM317 thread that flat impedance Vs frequency might be more a determinant of sound. Noise also would be the major worry with batteries - I've no way of measuring this - any suggestions?

Edit: Yes Salas, I was going to start with V1 but I need to re-read Tham's posts when I start the build & if V1.5 isn't much more complicated then that will be the one

I don't mean to be putting a damper on your reg by doing this just evaluating the various options. And with batteries there's tha added complexity of monitoring low voltage & chargers, etc.

Thank you for being so gracious in considering these - many designers have a "my baby is the bestest ever" attitude
 
Last edited:

iko

Ex-Moderator
Joined 2008
There are two simple tests for the output impedance. You won't get any journal papers published with either of them, but it will allow you to compare two different sources. You'll need a sine generator for this. There are several sine signal generator solutions freely downloadable if you don't have a HW one. You also need a high current n-channel mosfet, which will be the load. One 10k resistor across G-D and a variable resistor of 10k across the G-S pins. Set the variable resistor to about 500R initially. Apply the sine signal through a capacitor of about 4uF to the gate of the mosfet. Then connect the drain of the mosfet to the regulated output, and the source of the mosfet to ground. As you increase the resistance of the variable resistor across G-S, the mosfet will pass more and more current. I usually don't apply the sine signal until I get the current I want through the mosfet. The mosfet should have a good heat sink.

Then you use the AC setting of your DMM to measure the regulated output. I find it best to look at it with the oscilloscope. The smaller the AC on the output, the lower the output impedance is. I do this usually for a few frequencies of interest: 500Hz, 1kHz, 3kHz, 5kHz, 10kHz, 15kHz, 20kHz.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.