Looking for some feedback/suggestions on my scheme for regulating amp rails and voltages for use with OP Amp IC's. The current needed for the IC's is less than 30mA. The use of the MC78 and MC79 are for both buffer and also that I can easily swap them for even lower voltage output units for use with a wider range of LDO that follow them. In actuality, both U2 and U4 will be on a separate PCB that will be fed from U1 and U3.
Thx
S
Thx
S
If you really only need 30 ma then why not use a shunt regulator and series resistor.
If the -/+70V are from a conventional PSU (i.e. a 50-0-50 transformer) then you could look at using auxiliary rails derived from a capacitor dropper from the AC windings. That would be 'watt-less' (no heat) and you could then use that to feed a linear regulator.
If the -/+70V are from a conventional PSU (i.e. a 50-0-50 transformer) then you could look at using auxiliary rails derived from a capacitor dropper from the AC windings. That would be 'watt-less' (no heat) and you could then use that to feed a linear regulator.
Peavey uses a couple of 3 W 15 v zener diodes, preceded by suitable resistors to limit the current for 1.5 W. From +-85 v rails. Then a couple of capacitors. Dead simple. Look at PV-1.3k schematic on eserviceinfo.com
Of course their hD is only <0.1%, not .0003% like people like to brag about on solid state. My $700 speaker hd is 20 db down at all frequencies 54-14.5khz @ 1w, say 10%?
Of course their hD is only <0.1%, not .0003% like people like to brag about on solid state. My $700 speaker hd is 20 db down at all frequencies 54-14.5khz @ 1w, say 10%?
Last edited:
Technically I only need about 15mA, it's for an INA1651 that lists at about 11mA.
Wattage should be about a half, to one watt under normal running conditions.
I figure though that if I'm going through with it, that it may as well have some headroom.
Thanks for all the replies.. I've thought of a few different approaches, winding my own, adding another transformer/supply etc.. but I figure with such a low draw, and another ground issue to deal with, that it would just be too much for the application.
I'll investigate more of the ideas proposed.
Based on the attached scheme, would it be feasable to incorporate a limiting resistor (R1 and R2) without running into any issues? This resistor would double as a way of detaching or attaching the regulator system at other unknown times. If I were to do the math, I would just do the drop of 33v in fears of limiting too much. What would the proper way to do this be?
Thanks again
S
Wattage should be about a half, to one watt under normal running conditions.
I figure though that if I'm going through with it, that it may as well have some headroom.
Thanks for all the replies.. I've thought of a few different approaches, winding my own, adding another transformer/supply etc.. but I figure with such a low draw, and another ground issue to deal with, that it would just be too much for the application.
I'll investigate more of the ideas proposed.
Based on the attached scheme, would it be feasable to incorporate a limiting resistor (R1 and R2) without running into any issues? This resistor would double as a way of detaching or attaching the regulator system at other unknown times. If I were to do the math, I would just do the drop of 33v in fears of limiting too much. What would the proper way to do this be?
Thanks again
S
I did this with a chip amp which was sort of like Bob Cordell's design.
Wind perhaps ten turns and see what voltage you get -- then you can derive the appropriate turns ratio.
The mains are wired for soft start, yet I am looking to insert limiting resistors as it is..
I intend to use transformers from toroidy and they appear to be potted. I haven't yet inquired, however, there is a possibility to have an extra set of secondaries wired in with it if they do such. It's all about space constraints and I know that their standard audio grade 400VA will just fit as it is. I suppose I could drill a hole through the potting for feeding some turns through it. Can't be all that much to get 100mA. Little bit of solid 22ga about all I bet.
S
I intend to use transformers from toroidy and they appear to be potted. I haven't yet inquired, however, there is a possibility to have an extra set of secondaries wired in with it if they do such. It's all about space constraints and I know that their standard audio grade 400VA will just fit as it is. I suppose I could drill a hole through the potting for feeding some turns through it. Can't be all that much to get 100mA. Little bit of solid 22ga about all I bet.
S
Thanks for the replies and ideas..
I think what I'm looking at is something more like the following...
Simpler with fewer parts, better components, safer etc...
If I need a higher voltage than the fixed reg models provide, I can use the adjustable in their places.
This circuit has better built in current limitations using just the LT regs, whereas the previous could potentially draw too much current, even at quiescent state.
Even if I didn't use the regulators in my system and they were installed permanently, using this scheme, the power draw is so low in an unused state that I wouldn't even think about it.
If I need another regulator, it would be another LT reg on another PCB following these.
S
I think what I'm looking at is something more like the following...
Simpler with fewer parts, better components, safer etc...
If I need a higher voltage than the fixed reg models provide, I can use the adjustable in their places.
This circuit has better built in current limitations using just the LT regs, whereas the previous could potentially draw too much current, even at quiescent state.
Even if I didn't use the regulators in my system and they were installed permanently, using this scheme, the power draw is so low in an unused state that I wouldn't even think about it.
If I need another regulator, it would be another LT reg on another PCB following these.
S
Take a look at the LT3080 datasheet -- the device can be protected as 70 Vin to 15 Vout with the addition of a mosfet and zener -- the same should apply to the LT3091 negative regulator. This "newish" series of regulators uses a single resistor to set the output voltage, they are very low noise, high PSRR.
TI still makes the TL783 high voltage regulator, but there is no negative voltage equivalent.
TI still makes the TL783 high voltage regulator, but there is no negative voltage equivalent.
Good idea. May require a post-regulator, you won't know until you try. Without a lot of test equipment the easiest way to find out is fire it up and sweep the a.m. broadcast band for noise.
Read the datasheet; plenty good regulation specs, low ripple for Op-Amp +/- vcc . Maybe throw in a 10uF and .01uF to GND on each. An AM radio? Really? Check ebay for a cheap used Velleman PCSGU250. Hook it up to an old PC or laptop and you got a 2 Channel 12Mhz scope AND a functon generator that sweeps.. perfect for basic audio work and more!
Recommended by Analog Devices. It works.
My experience with DC-DC converters of this type (Murata and others) is that the switching transients can fritz up BJT junctions by "unexpected rectification" -- I didn't invent the term, ADI did.
You also might want to probe the input of the DC-DC converter as they often spew backwards. Put a probe on it and examine with a spectrum analyzer.
Not taking a shot at TDK by any means, they are an estimable company.
Thanks again
I am actually attempting to steer around using another power supply or another switching device since I have separate zobels of each analog ground and digital ground to earth ground, isolating them as best as I can. Originally I had an additional power supply +-15 for these aux devices, but I could not get around having to tie the grounds together and lose that specific advantage. Currently I use input transformers to bridge and isolate both sides, the raspberry Pi and it's power, and the amp and it's power. Using the above method allows me to keep the power supplies isolated and utilize just the analog side for the analog devices and the digital on their own. It's kind of a big deal actually because using the transformers (differential outputs/inputs) has achieved such good results that I now attribute much of the gain in reduction of noise in audio to be this particular thing. My goal is to do my best to provide another option in hopes that it will also provide a bridge from the SMPS DAC side of things to the quiet and sensitive linear. Isolating each power system is essential.
About the LT 3080 and LT3091, I haven't had the chance to look at them, it sounds promising although it appears to be something similar to what I am doing already.. I will check it out when I get a chance.
Thanks
S
I am actually attempting to steer around using another power supply or another switching device since I have separate zobels of each analog ground and digital ground to earth ground, isolating them as best as I can. Originally I had an additional power supply +-15 for these aux devices, but I could not get around having to tie the grounds together and lose that specific advantage. Currently I use input transformers to bridge and isolate both sides, the raspberry Pi and it's power, and the amp and it's power. Using the above method allows me to keep the power supplies isolated and utilize just the analog side for the analog devices and the digital on their own. It's kind of a big deal actually because using the transformers (differential outputs/inputs) has achieved such good results that I now attribute much of the gain in reduction of noise in audio to be this particular thing. My goal is to do my best to provide another option in hopes that it will also provide a bridge from the SMPS DAC side of things to the quiet and sensitive linear. Isolating each power system is essential.
About the LT 3080 and LT3091, I haven't had the chance to look at them, it sounds promising although it appears to be something similar to what I am doing already.. I will check it out when I get a chance.
Thanks
S
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.