I'm looking for the best design to implement a SLA-based power supply for a turntable motor. I have a 12v SLA battery, and a 12v brushed Maxon motor that I want to power with this battery. I want to get the best regulation possible, and hopefully keep the circuit simple.
I built a simple regulator using a 3-pin fixed reg, but encountered a slow, and steady fluctuation, which could be heard as wow in the music. Now I'm considering a discrete design using the very best low-tempco components possible.
I was considering a shunt regulator design like this one:
http://www.tnt-audio.com/clinica/regulators_noise4_e.html
I'm not a circuit designer, and are not pretending to be one, so I'm not really able to evaluate all the circuits and choose the best one.
Any ideas?
I built a simple regulator using a 3-pin fixed reg, but encountered a slow, and steady fluctuation, which could be heard as wow in the music. Now I'm considering a discrete design using the very best low-tempco components possible.
I was considering a shunt regulator design like this one:
http://www.tnt-audio.com/clinica/regulators_noise4_e.html
I'm not a circuit designer, and are not pretending to be one, so I'm not really able to evaluate all the circuits and choose the best one.
Any ideas?
With your 3-pin fixed linear regulator circuit, what voltage were you trying to supply to the motor? (And how much current does it need?) Can you post the schematic?
Since you mentioned a 12V battery and a 12V motor, I have to mention that any voltage regulator will need its input voltage to be higher than its output voltage (by more than its specified "dropout voltage"), for it to be able to stay in regulation. Even some of the low-dropout (LDO) types, such as linear.com's LT1083/84/85/86 family, need at least a 1.5V or higher difference between input and output (or somewhat less for less than maximum spec'd output current), although there are some LDO types that do have much lower dropout specs, down to less than 0.5V, I think.
By the way, when using a three-terminal regulator, you can get much lower output noise+ripple by using one of the adjustable types, WITH a bypass capacitor on its adjust pin. See, for example, the circuits in the datasheets for the LM317/LM337/LM117 and LM338, at national.com, and for the LT1083/LT1084/LT1085/LT1086 series (which are very good), at linear.com .
If you do need to construct a very-high-performance discrete voltage regulator, you might want to do some searches for Sulzer and/or Jung/Didden regulators. There are several threads about them, here at diyaudio.com. There should be boards and/or kits available for them. Or, for your application, you could probably just pick any suitable series or shunt type of discrete regulator circuit. They should basically all be _much_ better than a fixed three-terminal regulator, especially at higher frequencies. Just remember that some types have relatively-high "dropout voltage" specs, which in some cases might even mean that you'd need to use two batteries in series (or a higher battery voltage).
After looking at those plots of battery noise, at the link you gave, I think that you might also want to consider using a CLC "Pi"-layout filter, possibly with an adjustable LDO regulator after it. A couple of fairly-large low-ESR electrolytics to ground (maybe 2200uF or 3300uF low-ESR types, such as Nichicon UHE; see mouser.com), with a high-current toroidal inductor between them, in the current path (maybe a 10uH or more J W Miller toroidal; see mouser.com or digikey.com), should remove virtually all high-frequency noise. And the adjustable linear regulator should take care of any lower-frequency fluctuations. Be sure you use a bypass cap on the adjustable regulator's adjust pin, per the datasheets, which usually recommend something like 22uF (probably a low-ESR type) for the adj pin and 100 uF on the output (probably NOT a low-ESR type; low ESR there might cause stability problems). If you want more output capacitance, go much higher. I've seen reports of stability problems with intermediate values (between, say, 100uF and 1000uF).
Good luck.
- Tom Gootee
http://www.fullnet.com/~tomg/index.html
Since you mentioned a 12V battery and a 12V motor, I have to mention that any voltage regulator will need its input voltage to be higher than its output voltage (by more than its specified "dropout voltage"), for it to be able to stay in regulation. Even some of the low-dropout (LDO) types, such as linear.com's LT1083/84/85/86 family, need at least a 1.5V or higher difference between input and output (or somewhat less for less than maximum spec'd output current), although there are some LDO types that do have much lower dropout specs, down to less than 0.5V, I think.
By the way, when using a three-terminal regulator, you can get much lower output noise+ripple by using one of the adjustable types, WITH a bypass capacitor on its adjust pin. See, for example, the circuits in the datasheets for the LM317/LM337/LM117 and LM338, at national.com, and for the LT1083/LT1084/LT1085/LT1086 series (which are very good), at linear.com .
If you do need to construct a very-high-performance discrete voltage regulator, you might want to do some searches for Sulzer and/or Jung/Didden regulators. There are several threads about them, here at diyaudio.com. There should be boards and/or kits available for them. Or, for your application, you could probably just pick any suitable series or shunt type of discrete regulator circuit. They should basically all be _much_ better than a fixed three-terminal regulator, especially at higher frequencies. Just remember that some types have relatively-high "dropout voltage" specs, which in some cases might even mean that you'd need to use two batteries in series (or a higher battery voltage).
After looking at those plots of battery noise, at the link you gave, I think that you might also want to consider using a CLC "Pi"-layout filter, possibly with an adjustable LDO regulator after it. A couple of fairly-large low-ESR electrolytics to ground (maybe 2200uF or 3300uF low-ESR types, such as Nichicon UHE; see mouser.com), with a high-current toroidal inductor between them, in the current path (maybe a 10uH or more J W Miller toroidal; see mouser.com or digikey.com), should remove virtually all high-frequency noise. And the adjustable linear regulator should take care of any lower-frequency fluctuations. Be sure you use a bypass cap on the adjustable regulator's adjust pin, per the datasheets, which usually recommend something like 22uF (probably a low-ESR type) for the adj pin and 100 uF on the output (probably NOT a low-ESR type; low ESR there might cause stability problems). If you want more output capacitance, go much higher. I've seen reports of stability problems with intermediate values (between, say, 100uF and 1000uF).
Good luck.
- Tom Gootee
http://www.fullnet.com/~tomg/index.html
this sounds like a fault in the regulation.Capt Zach said:
can you measure the input voltage and the output voltage of the regulator when the wow is present and when the turntable is idling without playing.
The variable drag from the stylus may be causing the wow and the regulator may be "hunting" trying to stabilise the voltage, particularly if the motor/regulator does not have sufficient power to "pull the record" through the high drag parts of the music.
This may be a low frequency instability (often referred to as motorboating in audio) that could be cured by sorting the regulation not by changing the regulator.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.