Noted. It's either longer wires for the DC or the signal...but it sounds nice 🙂
Next time, maybe a design with the shunt integrated 😉
Next time, maybe a design with the shunt integrated 😉
Quick question. Revealing my ignorance, is there an easy way to measure actual vs programmed (bias) current draw? If I wanted to gauge the headroom.
Yes, that's the bias, so far as I understand. How about determining the load, how much current the DCG3 draws? Any simple way for that? Apologies if i am being dim. The DCG3 is biased at 123mA, so likely draws just under 150mA. Guess I can go with that.
Last edited:
aha, better formulated question, so you'll get proper answer
without knowing actual schematic of stage you're feeding, I can say that only way is either inserting small value resistor and measuring voltage sag, or inserting current meter
inserting - in between shunt reg and loading stage
without knowing actual schematic of stage you're feeding, I can say that only way is either inserting small value resistor and measuring voltage sag, or inserting current meter
inserting - in between shunt reg and loading stage
Got it. Thank you! Done.
The DCG3 draws ~133mA on the negative rails and ~140mA on the positive rails. So, if for the sake of heat management, if I target 20-25% overhead that would be about 170-175mA bias for the shunts, or 3.7R-3.8R for R3||R4 and R5/R6. Maybe use 7.5R for those four resistors. If I have this right, it would drop the overhead from about 90-100mA to 30-40mA. Does this seem good?
The DCG3 draws ~133mA on the negative rails and ~140mA on the positive rails. So, if for the sake of heat management, if I target 20-25% overhead that would be about 170-175mA bias for the shunts, or 3.7R-3.8R for R3||R4 and R5/R6. Maybe use 7.5R for those four resistors. If I have this right, it would drop the overhead from about 90-100mA to 30-40mA. Does this seem good?
overhead number is matter of eyeball, not some exact formula
whatever you are content with
point is that you don't need insane heat (huge overhead) to get proper reg performance - "just" loading current increased for few mA needed for reg function, then some
my eyeball for your 140mA load would be programming for 180mA
whatever you are content with
point is that you don't need insane heat (huge overhead) to get proper reg performance - "just" loading current increased for few mA needed for reg function, then some
my eyeball for your 140mA load would be programming for 180mA
Thanks Vunce. I had a single Sparkos +/-15v PSU before the GG's. Going from one to two PSU's is probably not a fair comparison, but I prefer the GG's. Likely placebo, but it seems the stage is wider/deeper, with the sounds coming less from the speakers. The very low end also seems better. I have a vivid imagination 😉 and will be making more of these.
Trying not to be too greedy ... and if understanding the posts above ... for targeting 30-32V, it seems changing R7 alone may not be ideal, and that R8 and R9 would need tweaking too?
let's earn this ......... 
post schematic with changes you're predicting, and I'll give you hint or two
you passed point of building just to have it ...........
post schematic with changes you're predicting, and I'll give you hint or two
you passed point of building just to have it ...........
Can I 'buy' a clue? Starring at the Shunty cookbook but can't figure out the current through U1 in GG. I suspect JFET3/R24 handles it but don't quite grasp it. Yet.
🙏
🙏
So, pure guesswork on my part. Trained in bio-stuff not EE 😖
Voltage of the shunt set by:
Where Iref is from red square:
Apologies for cluttering the otherwise nice GG thread.
Voltage of the shunt set by:
Where Iref is from red square:
Apologies for cluttering the otherwise nice GG thread.
there is no need to fiddle with voltage reference
first thing first
what's load current, rail voltages and what programmed current is?
first thing first
what's load current, rail voltages and what programmed current is?
This GG would be for the D1 I/V. So, load ~40mA and rail voltages +/-30V.
So, I would think programming the GG to 60-70mA would be plenty, say 10R for R3||R4 / R5||R6?
For the 30V, I am looking at increasing R7 but can't get the math right.
Lastly, C1-C4 would increase to 50V or more.
So, I would think programming the GG to 60-70mA would be plenty, say 10R for R3||R4 / R5||R6?
For the 30V, I am looking at increasing R7 but can't get the math right.
Lastly, C1-C4 would increase to 50V or more.