What frequency is the 36mV ripple? Is it there on the output even when the reg isn't connected to that load? How long is the cabling to the load? Does it have any local decoupling small caps? Other Ubib builds work in the particular application?
P.S.
Much or suboptimal spare current can upset matters if delicate. Play with the CC allowance for experiment. 1.5 Ohm first, then 1 Ohm R1.
P.S.
Much or suboptimal spare current can upset matters if delicate. Play with the CC allowance for experiment. 1.5 Ohm first, then 1 Ohm R1.
Freq was 30hz, rearranged my breadboard and now 3mv. Using 1.2 ohm resistor now at 500ma. R9 is 1K
Everything working now, time to get busy and put this in a chassis and clean things up some. Going to increase xfrmr to at least 80va, I think the 25va is a little small and may be making the BIB sensitive to changes.
Everything working now, time to get busy and put this in a chassis and clean things up some. Going to increase xfrmr to at least 80va, I think the 25va is a little small and may be making the BIB sensitive to changes.
Hmm the sub mains frequency and the drastic value change with breadboard rearrangement rather hints to interference. Do check again in the end. Did you measure the actual load's consumption?
When probing on some weird frequency indications, a crude verification method is power off and see if still there picked from somewhere. Common mode noise couples to non differential measurement loops. Don't forget to use the short ground coil on the probe's nose than the crock wire, but you already know that.
When probing on some weird frequency indications, a crude verification method is power off and see if still there picked from somewhere. Common mode noise couples to non differential measurement loops. Don't forget to use the short ground coil on the probe's nose than the crock wire, but you already know that.
A question concerning those who are using the SALAS board for Soekris DACs:
There are small inductors at the inputs for the power supplies.
What does the regulator think of having an inductor in the path?
It would be easy enough to bypass these if it would be better.
Any thoughts and recommendations?
Thanks,
There are small inductors at the inputs for the power supplies.
What does the regulator think of having an inductor in the path?
It would be easy enough to bypass these if it would be better.
Any thoughts and recommendations?
Thanks,
I had tested it with long lab leads which represent some inductance besides some resistance and it was fine. To check for specific situations use the scope. If the rails are calm straight lines in AC coupling 5mV vertical & 50uS horizontal settings everything should be fine. Use the short earth (coil like) probe's nose accessory.
Prefer the more stable technical result. If the rail shows as stable with input coil as with bypassed input coil, then its down to subjective preference keeping it in circuit or not.
Prefer the more stable technical result. If the rail shows as stable with input coil as with bypassed input coil, then its down to subjective preference keeping it in circuit or not.
Hi Salas,
I would like to know for what applications a shunt regulator like the UltraBib 1.3 is best suited for. And what is it about these shunt regulators as compared to the three legged ones that make them sound so special.
I am considering building Pete Millett's Nutube balanced line amp and was wondering if the UltraBib 1.3 would be a good fit here.
Nutube Balanced Amp
A +/-15v supply the board draws about 15mA from the +15V supply, and 75mA from the -15V supply (which also powers the filament). This is for each channel, so I would go with two of them.
Appreciate your input.
Thanks. nash
I would like to know for what applications a shunt regulator like the UltraBib 1.3 is best suited for. And what is it about these shunt regulators as compared to the three legged ones that make them sound so special.
I am considering building Pete Millett's Nutube balanced line amp and was wondering if the UltraBib 1.3 would be a good fit here.
Nutube Balanced Amp
A +/-15v supply the board draws about 15mA from the +15V supply, and 75mA from the -15V supply (which also powers the filament). This is for each channel, so I would go with two of them.
Appreciate your input.
Thanks. nash
CCS fed shunts (AKA super shunts) like the UBiB, are easier to use with relatively steady current drawing low to medium power clients. So the CC limit to load current ratio is predictable and moderate for good settings regarding spare current, stability, and dissipation.
Shunts both give and soak up power when series only give. Such shunts refer power to the output node. The main CCS isolates things located before their parallel voltage circuit. Excess current is constantly burned at the output so they are surely not "green". To the contrary. Loosely described sometimes as the Class A amps of the PSUs world. Those are general characteristics, their special qualities vary per design.
So they differ in architecture to normal three pin regs. Consequently they have their flavor and their followers.
All client circuits within the voltage and current envelope described in the build guide are UBiB powered candidates. I would set 125mA CC in the positive and 200mA in the negative for your example if double mono. For common between channels PSUs I would set 150mA CC pos 270mA CC neg.
Shunts both give and soak up power when series only give. Such shunts refer power to the output node. The main CCS isolates things located before their parallel voltage circuit. Excess current is constantly burned at the output so they are surely not "green". To the contrary. Loosely described sometimes as the Class A amps of the PSUs world. Those are general characteristics, their special qualities vary per design.
So they differ in architecture to normal three pin regs. Consequently they have their flavor and their followers.
All client circuits within the voltage and current envelope described in the build guide are UBiB powered candidates. I would set 125mA CC in the positive and 200mA in the negative for your example if double mono. For common between channels PSUs I would set 150mA CC pos 270mA CC neg.
Thanks for the straightforward explanation. Yes, I am a follower and love the flavor!
25 or 50VA transformers for dual mono? Antek seems to make the shielded AS series only for the 50VA. Dont know why.
In the example that I am considering, the +15v feeds the Jfet followers and the anode plate but the -15v is lowered to 5v by a LM317MTG to feed the grid and then another LM317MTG lowers the voltage further to feed the heaters. Do these subsequent LDO regs constrict the flavor of the shunt PSU?. Or would dropping resistors be a better choice?
Nash
Buy the 50VA ones IMO, they are better made and not expensive.
The shunts usually show some of their traits as pre-regulators too so listen first and then modify the negative rail internal arrangements if you feel its a must.
But ask the designer of that preamp circuit if he agrees and on how to do it best with passives or differently.
The grid bias feed (-10V) arrangement should be the main quality consideration I suppose.
The shunts usually show some of their traits as pre-regulators too so listen first and then modify the negative rail internal arrangements if you feel its a must.
But ask the designer of that preamp circuit if he agrees and on how to do it best with passives or differently.
The grid bias feed (-10V) arrangement should be the main quality consideration I suppose.