Power output of the low budget 30v/10a lab power supplies, and how they work with electric motors?

Other Hardware General Electronics
1

Hey friends,

For the background of my post: i am attemptind to build a test bench for fluid pumps powered by brushed dc motors. Up to now i have been using one of the well priced dc lab power supplies putting out 0-30v and 0-10a, adjustable.
Like this:

Now ive usually set it up with the amp limit turned to max, and then ran up the voltage bit by bit to pump start value, and certain voltage levels where i measure output at different pressures.

At this point i am trying to enhance the system to a point where i can control power supply, measure flow, measure and control pressure, and protocol the measurements to SD card.

I then bought a pretty costy dc power supply that will deliver true smooth DC from 0-12v.

At this point i noticed something: a motor that would start runnung at 0.7V on the lab supply would stay stuck to about 3.4V on the expensive unit, and then jump to fairly high rpm. I can then power down to about 1.6v before the motor stops. (The theory behind this behavior is clear to me).
This shows clearly that the lab unit is putting out a probably somewhat voltage-regulated signal that is pwm-tuned to cover lower voltages than a LM317 or similar just cannot reach. Apparently there is no or insufficient signal smoothing on the outlets. I dont have an oscilloscope, so i cant test the theory or check real output values.....maybe someone has done so and can share what hes measuring?

Now to my intention and the related questions: i want to run 2 different "setups": in a first run i want to create a pure dc pump map, so straight through the regulated precision supply.

Now the complex part: i want to run the same routine on a pwm controlled signal, possibly on 2 different frequencies. For this i want to put a RFP30N06LE (60v/30a) into the line, put it 100% high for the dc run, and then use it to pwm the signal.....

Now the lab unit does not appear to have a flyback diode. How does that work? Also, does the diode not initiate a short-circuit braking inbetween the cycles, and thus slows down the motor?

Thanks for inputs, ideas, comments......
Cheers
Hank

2

The user has to add such a diode to every switched inductive load. Batteries as power supplies also don't have flyback diodes...

That's not a braking effect, in contrast the diode stabilizes the continued current flow.

A scope is very helpful in diagnosing dynamic circuits.

3

An LM317 can regulate down to 1.2V according to its spec sheet, the characteristics of the pass transistors and their arrangement may also play a part. Your comments seem to suggest that the "pretty costly dc power supply" is a linear supply, but it is difficult to make any observations about a PSU we know nothing about.

Still, I am curious as to the reason for the different behavior.

4

@DrDiettrich
Interesting. Never had an issue without it over years. Maybe the cause can be found in the output behavior, but as you correctly stated: an oscilloscope would help a lot. But its a pretty steep invest for something i rarely can use....if at all its once a year.
Concerning the diode: in the diode thread i saw an 1N4004 being used, but its just 1A. If using a 650 class motor pulling 12-15A at 12V that will probably be insufficient? Maybe something like a 45V 20A rating would be more appropriate? How would the output behave with such a large diode if running a smaller 300 class motor on it?

@BitSeeker
https://www.ucarshop.com/product/ye-bg1537417

Thats the smaller brother, i have the 1kW version i think. Technology should be similar though.
Its either pwm with a very decent condensator, or its really true dc, but the start voltages are backed up when connected directly to 1/2/3 cell battery supply....

5

Ok, so that's a switch mode supply as well. Not likely to have LM317 or similar regulation.

6

Keep in mind that the startup current of a DC motor is 5 to 10 times higher than the rated running current.

What happens above is that the "expensive" supply is limiting the startup current to some low value, so the motor starts turning slowly. When the motor is finally at speed, the current limit is released to provide the normal running current.

1 Like
7

Right, the diode must support (at least) the motor current. Lower motor current is not a problem.

8

@jremington

Well if thats what the unit is trying then its trying for nothing. The pwm is starting much much smoother and slower. You can see voltage and amperage slowly rising, to the point where the generated field is strong enough to overcome rotor friction, and then it breaks free and runs fast. Thats the main reason i think it is very decently smoothed DC.
This is really one of the rare occasions where id love to have an oscilloscope to see whats going on. Ive seen some digital ones for around $100, but im not sure they have enough resolution to analyse the high frequency pwm....any reccomendation for a decent type oscilloscope that doesnt cost an arm and a leg?

Anyways first thing tomorrow will be ordering some adequate diodes and mosfets to do some trials.

9

In fact, that is a desirable characteristic of advanced motor drivers, called a "soft start".

Many good power supplies have adjustable current limits that can help protect external circuitry, if used properly. Check the user manual on yours for the details.

10

Yes. A soft start should make the motor start spinning early and slowly. Exactly like its happening with the pwm unit.

But with this i need 4x the voltage, before it nearly jumps off the table screaming at higher rpm....if thats a soft start feature then its useless, totally.

11

Ok i went to a friend and i found a totally awkward power curve.
The baseline is always at the voltage level i set. But i get a sine curve oscillation of plus minus 1.5v around the baseline, at a cycle time of 3ms. It reduces more and more in amplitude until it nearly levels at baseline. Every 80ms a new cycle starts again at full amplitude.
If i adjust base voltage on the supply then only the baseline moves up and down, pulling along the waves at identical amplitude and identical frequency.
Now who would design something with a strange output like that? Dos that make any sense to anyone?

IMG-20230102-WA0007
IMG-20230102-WA00071920×1080 75 KB

12

Is this measured with our without load?

At first glance it looks like some ringing of the output inductor but 3 ms period means about 300 Hz - a very low frequency for this. Maybe some problem with the control feedback loop?

EDIT: I see 12 cycles in each "step". 12 * 3 = 36 << 80. Either the cycles take longer or one step is only 40 ms. Assuming 6 ms cycles they have frequency about 150 Hz, that is close to 120 Hz, twice the mains frequency if you live in a strange part of the world.

EDIT2: I don't have experience with analog scope. When examinating the "screenshot" again it looks like there are many waves overlapping. Can you explain what we see? What time per division you used?

13

Hi,

What do you mean by power curve?

What are we actually looking at?
What is the load?
Where was the gnd of the scope probe connected?
Where was the scope probe input connected?
Is the probe x10 probe, if so was it set on 1:1 or 10:1?
What is the vertical deflection set at?
What is the horizontal time base set at?

It looks like the sync is not set right and are seeing a number of scans over writing each other.

Can we please have a circuit diagram?
An image of a hand drawn schematic will be fine, include ALL power supplies, component names and pin labels.

Show us on the circuit diagram where you connected the scope probes.

Also some images of your project so we can see your component layout.

Can you please post a link to spec/data of your motor?

Thanks.. Tom.. :smiley: :+1: :coffee: :coffee: :coffee: :australia:
PS. Sorry, I know this is a lot, but you are now in trouble shooting mode and this is essential information to help us see what is going on.

14

Remember, the PSU is supposed to be a current limited supply. I suggest we are looking at the results of the current limiting circuit either not liking no load or the result of how it deals with over loading.

15

The setup is very very basic, i dont think a sketch is needed.
The power supply is the cheap china 30v/10a PSU (see below).
The load that is connected is a mabuchi 370 motor (speed 400 class), plus to plus, minus to minus,directly, running without load.
The 2 osciloscope channel connectors are directly connected to motor + and -, set to 1:1 rate.
Voltages used were all between 1v and 5v.

Big correction: units is microseconds, not milliseconds.

Vertical i set 1v/cm, horizontal 10 microseconds per cm.
I think only the 2 main waves left and right are main waves, center appears to be a faded shifted wave,but im not sure.
I have never worked with an osci before, especially one thats older than i am, and the owner has not used it for decades either so he was no real help.....so part of the problem may have been me. But just getting to that point cost me about an hour.

IMG-20230102-WA0004
IMG-20230102-WA00041836×3264 181 KB

16

No problem. I try to answer as best as i can and am grateful for all infos or help. If i skipped something im sorry, just tell me and ill get it eventually :see_no_evil:

17

Hi,
So it is a BRUSHED motor you are using, what you are seeing is probably the ringing of the armature as the brushes traverse the commutator gaps, not uncommon.
Solution, 0.1uF or even 1.0uF capacitor across the motor terminals.
Not a power supply problem.

What is the spec of the motor, Volts amps etc.
Can you please post a link to spec/data of your motor?
Or at least the full part number of the motor.

Thanks.. Tom.. :smiley: :+1: :coffee: :australia:

18

Hi tom

Its a RK-370SD-2870 motor. Theres a flyback diode and small capacitor in the output of the power supply i have since found out. Running without load gave a fairly smooth dc signal.

However the signal seen is not motor induced. If it were it would change in frequency in accordance to rpm. But it does not. Frequencies and amplitudes remain absolutely identical, the entire curve just moves up and down in voltage....

19

If you have a ferrite bead, try it over the + lead near the load.

20

This from the link;

7.2V DC Motor

Voltage Range (VDC): 4.5-9.6
Current (Amps): 0.34
No Load Speed (RPM): 16500
Rated Speed (RPM): 13790
Torque (g-cm): 60.9
Power (W):8.61
Terminal Type: solder
Shaft Diameter: 0.079"
Shaft Length: 0.47"
Size: 0.96"D x 1.21"L

What voltage and current limit do you have the power supply set at?

Try 5Volts and 1Amp current limit.

Can you show a scope trace with 5V supply?

Can you please post an image(s) of the scope setup?

Thanks.. Tom... :smiley: :+1: :coffee: :australia: