This may be a dumb question but.. when I checked the output of my DPS5015 there was 30mV ripple (@ 68Hz), that's with 15v nominal output and 2A. I tried placing a 470uf/63v capacitor across the output and the ripple disappeared to something negligable. Tried a 100uF and the ripple was only reduced to 25mV. Would it be OK to run it with the 470uF cap or is it going to burn my circuits when the power is switched off? I assume the stored charge has to go somewhere
(in the photo below the display refresh on the electronic load is slow, it actually reads 14.9?V and 2.00A)
A 470 µF is rather large, but not unusually large for a power supply of that power range. I would not expect the current limiting of this supply to be very fast either, so the extra charge from the 470 µF would not change that much. If at all the extra cap makes a difference on a short or on load transients, not during turn off.
A 68 kHz ripple frequency would be rather low for the switched mode supply. The load looks like an electronic load. If either the load or the supply is not really stable with difficult load, there is a chance the 68 kHz is due to the supply and load not working well together. There are chances that the load is the culprit.
Thanks! I'll try using a 50w power resistor as the load and see what happens, it whould be OK for a few minutes without overheating.
edit: result was the sme ripple frequency, larger ripple (35mV) and way reduced effect of the 470uF cap (down to about 20mV, so less than half reduction). The PSU output was 5A and lower voltage because the resistor is only 1 Ohm.
...way reduced effect of the 470uF cap (down to about 20mV, so less than half reduction). The PSU output was 5A
Not surprising the capacitor was too small with a 5A load. 5A will drain the capacitor in no time when the voltage dips.
I guess so. I will stay with the 470uF though, if I really need a 5A output I won't be reaching for this device
Hm, I ask myself, whether it wouldn't be possible to "abuse" such a module as an electronic load?
I have to discharge LiIon batteries from rc airplanes from time to time. If I'd attach these batteries on the input side of the module (as a source) - shouldn't it be possible to adjust the discharging - current on the output-side for a controlled discharging of the batteries? Burning the power shouldn't be a big issue, any piece of constantan, an old toaster or whatever should do.
This is maybe a weird idea, but in absence of a real electronic load, wouldn't t this be an alternative?
I did not like the cheap plastic look so i came up with this.
Very neat - except for the four mounting screws for each module. If you can't hide them, then they need to be a little more "industrial", IMHO. Maybe a small, knurled head?
But aside from that minor quibble, it is certainly appealing.
All it is missing is an earth terminal - and dual tracking capability.
I also just finished building a little enclosure around two DPS5005 modules.
I also just finished building a little enclosure around two DPS5005 modules.
Another nice job
Wish I could do the same. Be proud
HAH!! I have that same case ... and my vision is almost identical.
It's just waiting for a couple of those modules.
HAH!! I have that same case ... and my vision is almost identical.
It's just waiting for a couple of those modules.
Nice! What do you plan on using to provide the input power? I've currently got two ThinkPad laptop chargers in there.
Haven't planned that far ahead yet. Will be going through my box of transformers to see what I might have that will fit. If not, then it's onto the other boxes of stuff....
Having purchased several of the low cost power supply modules and with a bunch of extra thumbwheel switches, I thought that I'd make my own version.
Attached is a picture of the voltage regulator/metering circuit which I just constructed. I used a LT3080 1.1 amp regulator for my first try. I might also build a 3 amp LT3083 version. The display shows the voltage in red and the current in blue. The side switch changes the voltage display between measuring the output and input voltages.
It doesn't have current limiting but does make it easier to change the desired voltage. It's somewhat a combination of old and more recent technology. It is just a spare time project to test the IC.
How did you mount in the case? ... Do you use one or two power supplies?
With the LT3080 regulator that I used rated for a maximum input/output voltage of 36 volts, I thought that I could live with the minor limitation of using the similar 33V 3A meters that a lot of people have on eBay:
https://www.ebay.com/itm/Dual-Display-4-Bit-Voltage-Current-Meter-DC-0-33V-1A-3A-Voltmeter-Ammeter/282275457510?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649 The trimmer resistor on those is used to calibrate the current reading. I used a Bud CU-793 enclosure. I used relatively primitive methods to cut rectangular holes for the meter and thumbwheel switches. Since I already have a bunch of regulated and unregulated supplies, I made it to be used with any single supply.
The voltage and current reading accuracy is mostly a function of the meter. I'm more concerned on how closely the output voltage to a precision meter compares to the setting on the thumbwheel switches. They select a single resistor which sets the output voltage. I generally used 0.1% resistors, sometimes paralleling the resistors to get the precise value desired. For example, I added a selected 100M ohm resistor in parallel with a 806k resistor to better achieve the desired 800k value. I have not yet completed the verification of that accuracy.
I have not yet completed the verification of that accuracy.
I just completed my verification on how well the LT3080's output voltage matched that chosen by the thumbwheel resistor selection. A typical output reading was under 0.3% off of the thumbwheel value. I suspect that the few outside of that range were due to my compromise made on the resistor value choice (such as using a 604k resistor when a 600k resistor was desired).
Attached is the accuracy data using my Agilent 34401A as the reference. The economy meter used had good accuracy as well. For the tenth of a volt accuracies, I calculated how far the increase in voltage compared with the voltage produced with the thumbwheel set for 10.0 volts. And with 32 volts input and the LT3080 set for a 1.0 volt, the output voltage was about 6.7% high. However, with an input voltage of 10 volts with that setting, the 1.0 volt output was just 0.03% low.
4) Yes, there is noise and ripple on the outputs, but nothing that an LC filter can't cure.
Moin German_EE, how much noise and ripple have you experienced?
With which model?
I've got a basic Sigrok driver written for these devices that I'll be submitting upstream shortly, but I've hit a minor sticking point.
The protocol has a MODEL register (0x0B) that identifies which device is connected, but the documentation doesn't provide any information about what each device actually has here. My DPH5005 returns 5205, so it's not a perfect mapping to the model number (although that may be just due to a name collision with the DPS5005.) I tried asking the company directly and hit a language barrier.
If anyone with a comms-enabled device other than the DPH5005 is in a position to read out the value from their unit and post it it'd be much appreciated.
(alternatively I could just write off-scale high values for V and I and read back the clamped values, but that's hardly elegant...)
Edit: I got in touch with the guy who translated the protocol doc into English and he confirmed that the register matches the model number for all units *except* the DPH5005, so I'm all set.