![]() ![]() ![]() (A friend has some audio circuits they'd like to play with, with similar voltage requirements.) I have microcontrollers and circuits that can measure both the voltage and the current at sufficiently high rates (~ 100 kHz sample rate) for me to detect stuff like current demand spikes, and the reaction of the power supply to that. Some background: I have a number of circuits, including Linux SBC:s, that take 5V or 12V at relatively high currents. This could be seen as a lab supply, but with a very limited voltage selection, and no constant current mode only current limiting. I would prefer it to be rather robust against stupidity, because I do stupid stuff when I'm still learning. An important factor for the circuit is the load capacitance: is it fixed, known, low ESR ? Another design point is if there is a current limit needed, either just for protection or as full lab supply with adjustable current ? For the positive side there are also a few high power LDOs that may be enough for 6 A. The LM723 is actually not that much more than a reference + an OP. For the negative side one may have to go for a classical regulator made from OP(s), reference and power transistor. Instead of a LDO, even a simple TL431 may be a better starting point for a high current regulator. LM723, LT175, some Micrel parts) However these are mainly available for the positive side. ![]() There are special regulator chips to work with an external power transistor (e.g. The parts are not that well defined to know upfront how bad the behavior would be before actually building it. 2 V) and stability / overshoot is really tricky. These circuits often have a high drop out (e.g. Using a LDO (with internal power stage) together with an external transistor to boost the power level is usually not a good idea. If the input part is a SMPS passive filtering with an inductor can be quite effective. So one would have to use a single power stage and maybe the capacitance multiplier only for the reference / regulation circuit itself, not the main current. Recommendations? Suggestions?įor the step from 7.5 or 8 V to 5.5V there is no really much headroom for a capacitance multiplier. With small changes (P↔N, opamp input range consideration), that should work for the negative rails too. Since I kinda want four of these (±4.5-5.5V 8A, ☑1-13V 6A), and I'll have accurate monitoring of the output (adjusted via a potentiometer), I'm thinking I might be better off doing discrete regulators based on a voltage reference (shunt, say TL431) suitably adjusted and filtered, with an opamp comparing the output to the voltage reference, and driving the base of a suitable pass Darlington transistor. Does this make any sense, or have I missed something important here? How could I implement the same for a negative rail? The supplies I have are fully isolated (class II, no capacitors on output to protective ground). R5, R6 form a voltage divider that control how Q3 reacts to voltage drop over R4, for foldback current limiting. While Q2 does not normally conduct, it has to be able to dissipate full load current during overloads. When the voltage drop over R4 is large enough, Q2 turns Q1 off. (I do believe an OnSemi FJA4213 or NJW0302G would be suitable here recommendations?) R2-R6, Q2, and Q3 implement overload protection. It should be a PNP transistor with low emitter-collector drop, in its active (non-saturated) region. When the current through R1 is high enough, the voltage drop over it turns Q1 on AoE uses 0.6 Ohm resistor here. R9 is the minimum load, if the regulator needs one. R7 and R8 limit the adjustable range to the designed safe and tested region (wrt. ![]() R7, RP1, R8 is the voltage divider for adjusting the regulator output. L1 is a ferrite bead to absorb any leftover very high frequency noise. R10, C1, R11, C2, R13, C3 form a three-stage low-pass RC filter, with total series resistance suitable for Q4 base current (say about 50-100mA for TIP142). Combining the advice from Art of Electronics, here's the circuit I've come up with: Q4 is a Darlington transistor, used as a capacitance multiplier. Slow turn on is not a problem, although I do want to avoid any voltage overshoot. MIC29752 comes close (7.5A), but for full 8A output, it seems I need an outboard transistor booster. Having a class II isolated 7.5V 8A power supply (adjustable up to 8V), I want to regulate it down to a nice, low-ripple 5 VDC –– say, adjustable between 4.5V and 5.5V, but at full 8A. power supply circuits is asking for your help. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |