Variable linear PSU and analogue meter maximum voltage.

[davelectronic]

I'm looking for some help with a linear power supply, with analogue panel meters. The power supply will go from 1.25 Volts up to 42.00 Volts DC. The Voltage panel meter is maximum deflection of 30.00 Volts DC. The PSU uses the LM317 HV variant, is there a way to limit the voltage potentiometer to 30.00 Volts DC ? The potentiometer is a panel mounted 10 turn type linear potentiometer. I want to do this so I don't damage the panel meters movement. Thanks for reading, any ideas help appreciated.

[Kleinstein]

One can limt the adjustment range by chosing the right resistance. Usually the LM317 circuit use a fixed 240 Ohms for the one side. This number is not fixed and one can do the fine adjustment there. The main purpose is to ensure the minimum current to get regulation.
The LM317 is not very good in giving up lots of heat. At higher voltage the SOA protection may turn in quite fast. So at low output voltages the ouput current may be limited if used from some 40 V on the input.


The usualy analog movements are not that sensitive and can stand a higher voltage than they can show on the scale. An factor of 1.5 over the range is usually not a problem.  To use the meter also for a higher voltage, one may add a 2nd range, like a factor or 2 or maybe 1.5 from an extra swtichable series resistor.

[davelectronic]

Thank you for the help. I didn't plan to draw to much current from this PSU, it was for testing other lite loaded circuits. So is, or can R1 usually 240 ohm's be a potentiometer for fine adjustment, while a potentiometer for R2 for course adjustment ? I didn't know about adding a switchable resistance in series with the voltage meter, so handy to know. I just didn't want the voltage wound up and damaging the meter.

[tunk]

can R1 usually 240 ohm's be a potentiometer for fine adjustment

No. You can either replace R2 with a multiturn potentiometer
or put a smaller potentiometer in series with R2. E.g. if R2
is 10k use a 1k potentiometer in series for fine adjustment.
You can adjust max voltage by replacing R1:
https://www.ti.com/lit/ds/symlink/lm317hv.pdf?ts=1638951107071&ref_url=https%253A%252F%252Fwww.google.com%252F

[tunk]

Add a resistor in series with a potentiometer

Are you thinking about R1 or R2?
If R2, then you will shift the adjustable range upwards.
If R1, the potentiometer will need to be lower, maybe around 100 ohms.

[davelectronic]

I was going with 120 ohm's for R1 it gives tighter regulation. And a 10 turn potentiometer at 5k ohm's. My transformers are 15 Volts AC each, I'm going to put them in series for 30 Volts AC. After rectification and filtering it's probably going to be over 40 Volts DC. It's the meter that only goes to 30 Volts DC, although I'm going to use the high voltage LM317 that takes a higher input voltage. And I'm going to use 6 X TIP147 transistors to get above the 1.5 Amp regulator rating.

[Vovk_Z]

Add a resistor in series with a potentiometer

Are you thinking about R1 or R2?
If R2, then you will shift the adjustable range upwards.
If R1, the potentiometer will need to be lower, maybe around 100 ohms.

Yes, my guess was wrong, it was the opposite.
VO = VREF (1 + R2 / R1). So it is easy to get R1 for a given potentiometer R2.
If the potentiometer is 5 k then R1 has to be about 220 R. Those values give us 29.6-29.7 VDC at the output.
If the potentiometer is 2 k then R1 has to be less than 100 Ohms (something like 87 R, it is not from E24), it is too low for me (but still can be used).
If the potentiometer is 10 k then R1 has to be 430 Ohms. It gives 30.3 VDC.

87 R is 100 R and 680 R in parallel. Or 120 and 330 R in parallel.

[Kleinstein]

I was going with 120 ohm's for R1 it gives tighter regulation. And a 10 turn potentiometer at 5k ohm's. My transformers are 15 Volts AC each, I'm going to put them in series for 30 Volts AC. After rectification and filtering it's probably going to be over 40 Volts DC. It's the meter that only goes to 30 Volts DC, although I'm going to use the high voltage LM317 that takes a higher input voltage. And I'm going to use 6 X TIP147 transistors to get above the 1.5 Amp regulator rating.

It is not so easy to add power transitors to increase the maximum current. There is a simple circuit for higher power in the DS, but this circuit compromises on the stability. So it may be just acceptable to supply a fixed circuit, it is not good for a standalone power supply where one does not know what will be connected. It can work most of the time, but still oscillate with a load that has too much capacitive load.

The no load voltage from a relatively small transformer can be quite a bit higher than nominal. So the 2x15 V AC may give you a bit over 50 V DC worst case with no / low load.  With load the minimum voltage would be around 35 V (depends on the size of the filter cap). So an upper limit if some 30 V is about right. More voltage would be available only with a large filter and low load.

[davelectronic]

Thank you for the explanation for R1 / R2 and potentiometers values. The circuit is from a reputable source, I have built circuits from this site, and had no problems, where they function as intended. Below is the circuit I was going to use for a variable high current power supply. Any added circuit protection will be added in addition to the main circuit.
Credit for circuit to Max from Idaho USA and project Icarus.
Also Andy Collinson circuit exchange.

[Kleinstein]

The cicuit has several problems. Not all are easy to fix, but one can at least avoid some of the problems:

For the TIP2955 0.1 Ohms is to low to get reliable current sharing - it may work with very good thermal contact, but may also fail. The TIP2955 should have more like 0.33 or 0.47 ohms each.

There is no sensible limit for the short circuit current. So a short would likely blow 1 or more of the transtors and than deliver the full voltage to the output til the resistor will go up in flames. A 30 A fuse would likely not blow in time. Some current limit would be a really good idea.

The 1 A fuse would likely never blow and if it would, it would more like make things worse as there would than could be still voltage at the output. With 100 Ohms there can be just enough voltage to turn on the transistors from the 5 mA the LM317 circuit draws. If at all reduce the 100 Ohms to more like 50 Ohms to get more current from the LM317 before the transistors contribute.

The 470 µF cap at the LM317 input side would cause quite some ripple from the raw supply to come though. So at higher current that can be very high output ripple. The 470 µF would be way to large - more like 470 nF as a suitable size.

The regulation would still have the 30 A fuse in series. The fuse would be better before the rectifier and than more like 15 A max. With some 35 V raw voltage the TIP2955 is not good for more than some 1.5 to 2 A each.

[davelectronic]

I'm sticking to the circuit loosely, but not exactly. I will use TIP147 transistors, I have several PSU's all similar hardware. I want a variable PSU and a fixed PSU from them. Choosing the LM317 will get me closer to 13.80 Volts DC at 30 Amps, what I need from one of them. The other I need variable for testing circuits for now. I don't have the funds for an expensive lab bench power supply unfortunately. Going to zero volts on the variable unit would be useful, I have seen two options to achieve this. The fixed PSU only needs to supply 10 to 15 Amps, very rarely 20 Amps. I just figured the extra head room would be safer in terms of power delivery. Your mention on reducing the 100 ohm base resistor to 50 ohm's is interesting one. The picture below uses the same circuit using a 7812 with 6 X TIP36C transistors. That's been in use 6 months or more now, but I wanted closer to 13.80 Volts. Another controversial is it uses two rewound microwave transformers in series for the secondary low voltage. Hopefully I can achieve these two PSU builds.

[Kleinstein]

A full lab supply circuit is not that much more expensive, though a bit more tricky in the developement. There are a few circuits discussed here in the forum, using a LM324 / LM358 for the control that are reasonable cheap to build. The expensive parts are the case, transformer and filter cap anyway.
The big plus is getting some short circuit protection, not to blow the ciruit and possibly later on the connected circuit.
IFAIK there is a way to add a crude fixed current limit also to the shown circuit.

[davelectronic]

I've seen circuits for home build lab bench power supply with a rotary switch for current limiting through fixed value resistors. That ideas is not one that appeals to me, I'd want continuously variable current limiting on a potentiometer if that could be achieved. I will look up the chips you have mentioned, I'm not familiar with those. I had looked for an LM317 bench lab power supply circuit, but can't seem to find one that can do this. Ideally I would like a 0 - 30 Volts at 10 Amps lab bench power supply, preferably linear supply. Sellers have lots of smps PSU's but very few linear models. The only linear affordable lab bench power supply I can find in budget is 0 - 20 Volts at 2 Amps linear supply. That's just to low on current and voltage.

[Kleinstein]

A linear power supply by design can worst case produce a lot of heat. For a DIY solution it is a very good idea to start small, with low power: the control parts is essentially the same, but if things get wrong less damage is done. One usually can not hat too many supplies, even if just small ones.
There are cheap linear supplies for some 30 V and 5 A claimed power, though those chinese ones may be a bit optimistic with the maximum current.
A DIY solution is usually not cheaper, unless you allready have some of the expensive parts, like the transformer, heat sink and case.
Already getting a DIY lab supply for lets say 25 V and 1 A is some achievement and good for learning before trying it large.

The LM317 is a voltage regulator and not a good starting point to build a lab supply. Having the power element and reference in the same chips is not a good start for a stable voltage, as the chip gets hot. So for a lab supply the way to go is with seprate reference, regulator (e.g. OPs as the active part) and output stage (e.g. transistors or mosfets).

A suitable relatively simple lab supply circuit is in this thread:
https://www.eevblog.com/forum/beginners/lm324-power-supply-with-variable-voltage-and-current/msg3745138/#msg3745138

[ledtester]

...
The picture below uses the same circuit using a 7812 with 6 X TIP36C transistors. That's been in use 6 months or more now, but I wanted closer to 13.80 Volts. Another controversial is it uses two rewound microwave transformers in series for the secondary low voltage. Hopefully I can achieve these two PSU builds.

You might want to take a look at the schematics for the Astron series of 13.8 VDC power supplies, e.g. the Astron 35A. They are proven designs and widely used to power radio equipment.

Schematic:

 astron_rs-35m_rs-35a.pdf (94.26 kB - downloaded 16 times.)

Repair video:

ASTRON RS-35A Power Supply Repair - Mr Carlson's Lab
https://youtu.be/dXtSTpBXItI

[davelectronic]

At higher currents with previous project PSU's, as long as I've used forced air cooling heat isn't to bad. It's definitely there, at 12.60 Volts DC 25 Amps continuous I've had readings on the semiconductors (not regulator) of 70 - 75 C° which I thought was exceptable for the load of halogen lamps it was powering. It's not my intended load, that's HF radio power use, so not continuous. I have just started Manhattan style PCB construction, as it offers higher current capabilities over copper strip board, which is difficult to accommodate high currents. Got to say why didn't I try this type of construction sooner, I only needed a bench shear to cut the copper clad board and strips. I have seen many of MR Carlson's lab videos, and have watched that PSU repair a while ago now. My phone (android) won't open the PDF file. But later I will look at it on the PC, thank you for the link. I have four PSU that don't have circuits yet, but ready to use once they have circuits in them. These are two cases bolted together to make room for bigger high current transformers. The other 4th PSU is in my kitchen awaiting further construction.

[BrokenYugo]

Looking at that schematic, some sort of soft start/inrush limiter would also be wise. 47,000uF is a big cap to charge in a hurry and for 30V 10A the minimum transformer VA is starting to get into the territory where it's recommended.

[davelectronic]

I've not used any inrush components in previous projects, but yes your right it would be a good idea for these higher currents and loads during switch on. The PSU in the first image has 4 X 47000uf capacitors, as it uses two microwave oven (MOT) transformers rewound for low voltage, and primary and secondary windings in series. A thermistor and mains filtering would be ok for inrush purposes ?

[davelectronic]

Thanks guys for all the help.