1

Briefly about my background, which might inform the question. I'm currently self-educating myself on some electrical engineering on-the-job. I'm learning from Art of Electronics and some other resources, and I studied physics in college.

The problem is this, I have several devices that I need to power from a single source. Their maximum power requirements are the following:

  1. 24VDC, 1A, 24W
  2. 24VDC, 1.35A, 30W max
  3. Three with 24VDC, 46mA each
  4. Two motors, 18VDC, 0.65A each

The total max current through my system is roughly 4.4A. I need to build a circuit that will handle power to all of these devices. My first assumption was that one should just be able to hook these all up in parallel across a 24V supply that will handle the current. What I have done, which is probably mistake number 1, is hook them all up directly to an 18V supply in parallel (the largest table-top supply we have in the lab). The result: something burned. I imagine if I had a 24V supply it would more than burn.

So, now I'm nose deep in Art of Electronics and these forums looking for principles, concepts, and precautions I should consider when trying to power such a system. I would greatly appreciate if you folks could point me in the right direction as I'm lost in the sea of technical literature.

Basically, I'm hoping someone will tell me "Hey, you need to build this standard current buffering circuit to prevent your parallel devices from drawing too much current and burning up the rest of your system" or "Go look up this concept that should lead you in the right direction as to what happens when you're drawing all kinds of currents in parallel like that"

Thanks in advance folks, hope I can be helped here.

lbman
  • 23
  • 1
  • 5
  • first off, they all need to accept the same input voltage. So you can't power those 18V motors without some kind of voltage regulator before them. – Hearth Feb 13 '19 at 23:15
  • I'm not sure what you mean by "current buffering circuit"... Are you wanting a current limiting circuit? – Hearth Feb 13 '19 at 23:16
  • If you are reducing the voltage by 25%, your current will increase 25%, so you probably weren't drawing 4.4 amps, more like 5.something, that is if the 24V devices can run from 18V at all (some can, but the amperage goes up to compensate for decreased voltage)... – Ron Beyer Feb 13 '19 at 23:18
  • Most devices work over a specified range of voltages, 18V is close enough to 24V that it's possible there's some intermediate voltage at which everything is going to function properly. Some devices can have problems with prolonged undervoltage. It might be easier to say if you told us what sort of devices these were. – Drew Macrae Feb 13 '19 at 23:18
  • Yea I do plan to have a voltage regulator in there for that but for now I've been using an 18V supply so the other components have been under-voltage. – lbman Feb 13 '19 at 23:18
  • @RonBeyer I though that decreasing the voltage would decrease current also. Assuming a naive V = IR, R is constant, V is lower and so I is lower. – lbman Feb 13 '19 at 23:21
  • Ohms law only applies to resistors. Most things aren't resistors, so I = V/R doesn't apply to them. – Simon B Feb 13 '19 at 23:24
  • @AndrewMacrae Yea I read that usually it's not a problem to go under-voltage and my devices do work with the lower voltage when powered individually. It's when I hook them all up at the same time that the issue arises. So my devices are 1. a pump running on some stepper motors 2. a device with what I think is a servo in it 3. devices, running through drivers, whose main components are magnetic coils that turn some mechanical stuff and 4. two stepper motors running through motor drivers – lbman Feb 13 '19 at 23:25
  • No, what you need to apply here is Watts = Volts * Amps. The device will take a constant wattage to run, (if) you can vary the volts, the amps will have to change to keep the wattage the same. Ohms law does not apply here because it is not a resistor. – Ron Beyer Feb 13 '19 at 23:25
  • 2
    You haven't actually said what the "something" was that burned. It's totally normal to just wire things in parallel, without any fancy current regulators. Are you sure it wasn't just some mistake, such as wiring something the wrong way round? – Simon B Feb 13 '19 at 23:25
  • 3
    By the way, Motors, can generate a significant back-feed voltage, which is why there are a lot of diodes used in motor control circuits. Don't be surprised if you unload your motor and you get 50+ volts feed back through the system. – Ron Beyer Feb 13 '19 at 23:26
  • @RonBeyer It's not a resistor but doesn't Ohms law apply to a circuit with a given impedance? So it could have resistors, capacitors, and inductors and still behave somewhat ohmic. I'm just trying to understand how applying less of an energy gradient can cause more current to flow. How can the circuit pull wattage? If there's no energy shouldn't it just not work? From the datasheet it also says that 30W is the max wattage so it should be variable. – lbman Feb 13 '19 at 23:39
  • @RonBeyer I'm sure you're right. Just trying to wrap my head around it. – lbman Feb 13 '19 at 23:41
  • @Ibman I was confused about different load types too, and eventually posted [this answer](https://electronics.stackexchange.com/a/91976/2028) once I'd clarified them through research. Check it out, it's very likely that other than the motors (inductive), you have mixed load types in each device. – JYelton Feb 13 '19 at 23:42
  • @JYelton I'm glad you brought up the various loads that can be applied. I'm working my way out of the dark ages with everyone's help here. I'm still not quite sure how a different load can create a power defined circuit though. Are the loads capacitive then? Trying to resist a change in required voltage by ramping up current? – lbman Feb 14 '19 at 00:05
  • I'd love to revisit the load types again, but honestly it's too much for comments. Check out related questions on the site, or ask a new question about that specifically. – JYelton Feb 14 '19 at 02:31

1 Answers1

1

The book you're studying is very good, however I am not sure it covers in a straightforward manner the parallel connection of different devices (where a device could contain many additional circuits).

Consider just the first five devices you mention:

  • Device 1: 24V 1A
  • Device 2: 24V 1.35A
  • Device 3-5: 24V 0.046A (ea.)

Since they require the same voltage, you can connect them in parallel. The power supply will need to be able to supply their sum total current requirements, ~2.5A (plus additional for inrush current).

If these devices are tolerant of lower voltage, they will try to pull more current to compensate (assuming they aren't just resistive loads like heating coils or something).

Recalculating the current to maintain power for the first five devices:

  • Device 1: 24W, @18V = 1.33A
  • Device 2: 32.4W, @18V = 1.8A
  • Device 3-5: 1.1W, @18V = 0.061A (ea.)

Adding the two motors:

  • Device 6-7: 11.7W, @18V = 0.65A (ea.)

Summing all the current requirements results in ~4.6A, which isn't much more than the 4.4A you originally cited as your total requirement.

I thought it might be helpful to go through and add up all of the device requirements, but in this case, I don't think the problem necessarily lies with the power supply (unless it cannot handle >4A) or voltage differences. Per Ron Beyer and Simon B's comments on your question, I'd suggest adding flyback diodes to the motors to protect the other devices, and double-check your wiring to ensure you haven't simply made a mistake.

JYelton
  • 32,302
  • 33
  • 134
  • 249
  • I see, this may be part of the problem then. In the datasheet for device 1 they specify that 1A should be supplied from a regulated power supply and it was a resistor in this device that got burned. The only strange thing is that this only happened when all the other devices were hooked up with it. – lbman Feb 13 '19 at 23:46
  • Now, for devices 6-7, I'm giving them 18V but actually they take whatever I give them up to 30V, so I'm not sure they're set up to draw power. The devices are stepper motors working through a driver that just does some logic. In this case, the elements are just resistive and the current will decrease with applied voltage and shouldn't negatively affect my circuit correct? – lbman Feb 13 '19 at 23:49
  • 1
    @Ibman If they're rated for 18V (and not e.g. 18~30V), do not give them more than 18V. Motors are resilient, but that doesn't mean they won't melt if you mistreat them too badly. – Hearth Feb 14 '19 at 00:10