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.
Hi,

I built the 5 to 15V version. I will be feeding 5 volts. The problem is that I'm not sure for my exact target current. The shunt will be used for supplying the DAC analog part. The chip is AK4396. On the datasheet it's said 32mA typical and 47mA max current. So how to make as constant voltage as possible? Increase output current using the load resistor?
 
Member
Joined 2009
Paid Member
Hi,

I built the 5 to 15V version. I will be feeding 5 volts. The problem is that I'm not sure for my exact target current. The shunt will be used for supplying the DAC analog part. The chip is AK4396. On the datasheet it's said 32mA typical and 47mA max current. So how to make as constant voltage as possible? Increase output current using the load resistor?

You can measure with the old regulator how much current draws so you know the current target, using Salas reg for better perfomance you must increase x2 or x3 the target current.
 
i just checked current needs of my buffalo with the old psu, with a simple LD1085 regulator and 20000uf filter, my ammeter signs 450mA so your advice is perfect.. i think that i screwed something on the reg.. even if i checked all the circuit 5 o 6 times... gotta start over from scratch with another board, that's diy after all! :)
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Hi,

I built the 5 to 15V version. I will be feeding 5 volts. The problem is that I'm not sure for my exact target current. The shunt will be used for supplying the DAC analog part. The chip is AK4396. On the datasheet it's said 32mA typical and 47mA max current. So how to make as constant voltage as possible? Increase output current using the load resistor?

Just use 4R3 R1 and the set current will be around 140mA, enough for your circuit and enough for the reg to work decently for low impedance, heat logical. You can verify that your circuit does not ask for more than 50mA with a DVM before you start, so to be sure.
 
The R dummy was set to draw around 200mA and testing resistor at the output was set to draw 60mA. The voltage is higher at iddle (like 8 volts, but at output current 60mA it gets to 5V) Is it normal to drop this way and drop more when the consommation gets a little higher?
 
i just return from a few days of being away from computer.

i see that there is some interest from the little experimentation i did. i must say that i m not entirely satisfied with what i came up with and some builders might not like it. so i ll continue to test parts and when i come up with something(i d like a bit more ''drive'') good i ll post a schematic. to be honest, some mosfets i tried did not have the Gm i thought they would have. i still encourage people to build it and test parts as i cannot test them all.

the build itself is either a sslv1.1 or 1.2 shunt circuit with X paralleled bf862(or other jfet) instead of the ''standard'' CCS. i built it as a negative reg because of the parts i found worthy of trying. so i think that a battery voltage of around 6V should be good. And because it is a neg reg, only one reg must be connected to each battery set.

i will also try a diy resistor with manganin of 200 ohm per meter for a small Vref values
 
Last edited:
i would but i feel that as it is right now, this concept is not worthy of getting too much attention. its behavior is a bit unpredictable soundwise, manganin not really available in small quantity as far as i know and this particular gauge is difficult to work with. so i prefer to be sure that people will have fun, in every sense of the word, and post a ''how to'' if i ever come up with something good enough. and in addition, i dont have measurement tools with me.

my interest in doing such a resistor first of all is to evaluate how detrimental were commonly found resistors to the sound in general
 
Last edited:
May you help me with my shunt? The output voltage stay on 4,59-4,60V and turning the pot doesn't affect it. Before this to happen, I had found out I had an error in my schematic. I had swapped the top BC550 with the 2SK170
http://i42.tinypic.com/33eo9jd.png. There was voltage regulation using the pot alright, but there was a voltage drop when changing the output load even if it was much bellow the dummy load.

The SK170 was heating very much in this case, depending on the dummy load. So I put both transistors back to the correct places and now there's no more voltage reg.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Verify your CCS current setting is enough to feed the load and allows some extra for the shunt to self feed too. At least +20mA. If that is not the problem check both Mosfets work by having 3.5-5V between their G & S pins on a meter. If no voltage, exchange. If healthy, exchange all TO-92s in case one of those is dead.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.