DPS5005 portable mini lab PSU

[plazma]

I started building a mini lab PSU of a DPS5005 module and an aluminium case. At the moment there is only a Dell laptop jack at the back. Bananas and a switch will be added. Maybe also a USB for controlling the PSU from a PC. I bought the comminucation version of the module.
I'll also add a big battery pack inside the case.

[plazma]

The module is amazingly good for the price.
There was only 25mV ripple with 5V 5A output. Probably much smaller when measured with minimum length leads.
I'll test the transient response as soon as I get access to a good load.
I tested with a Keitley lab PSU but it is known to undershoot with a large transient (like 0 --> 5A in my tests).

[plazma]

Standby current is 20-30mA depending on input voltage. I managed to fit that 20Wh battery pack inside the case.

[StillTrying]

"Standby current is 20-30mA depending on input voltage."

Not too bad considering the display.
What's the noise and ripple like at low current outputs, about 5V and 10-20mA.
Have you got any capacitance across the back of the output terminals.

[plazma]

"Standby current is 20-30mA depending on input voltage."

Not too bad considering the display.
What's the noise and ripple like at low current outputs, about 5V and 10-20mA.
Have you got any capacitance across the back of the output terminals.

The ripple goes down to few mV in low currents. No extra capacitance at output.

[floobydust]

I have a DPS3005 coming in, I think it's the same design just smaller MOSFET and diode.
I plan to add ferrite beads/common-mode chokes and extra capacitors to filter out the really high freq. RFI noise.
Most circuits I will power from it do not like 10's MHz switching harmonics.

[Alex Eisenhut]

Very nice, I have that module in my watch list. I like what you did with it, and Floobydust's LC + CM choke approach will help reduce the noise a lot. Y

[plazma]

After some use I really like it. Compact enough for desktop use. In the picture it works from the internal 11.1V 6600mAh battery pack. I can also use the laptop power plug on the left side of the PSU.


I ordered this cable for the opto-isolated USB option. I'll add it to the back panel for PC control.
http://www.ebay.com/itm/Micro-USB-5-Pin-Male-Plug-to-Female-Jack-Panel-Mount-Extention-Cable-Black-30cm/252869032494

[floobydust]

Check out the OpenDPS project  Johan has reverse-engineered DPS5005 and written Open Source firmware for it, including ESP8266 WiFi.

I'm not sure about the DPS5005 wireless option, I did not order the USB/Bluetooth comms option and the JST GH connector needed.

[plazma]

Here is a quick translation of the communication protocol document:
https://www.dropbox.com/s/y15lca5fh9mo6bh/DPS5005%20communication%20protocol%20-english%20google%20translate%20version%20%20V1.2.docx?dl=0

[firewalker]

I have ordered this module some time ago. I intent to use a micro with two individual rotary encoders, in order to set voltage and current values quickly (through the serial interface).

Alexander.

[Kalvin]

Have you checked how the DPS5005 behaves during power on / off and output enable / disable? I guess it is safe as you would have otherwise fried your circuit already. Would you happen to have any scope waveforms to share?

[plazma]

I have some transient scope pictures. I'm however waiting for a load to do more precise testing.

[Kalvin]

I have some transient scope pictures. I'm however waiting for a load to do more precise testing.

But your initial measurements indicate that the power supply does not overshoot when enabling/disabling the output or when the power supply is turned on/off?

[plazma]

I measured the transient with a sudden 5A load. On/off is yet to be measured. I'll measure it at the same time with the proper load. The unit is off by default when applying DC input.

[plazma]

There was large undershoot and small overshoot with 0A to 5A  to 0A sudden load change. The undershoot is also a feature of the PSU I used as a load. That is why I'll retest with proper load.

[Kalvin]

The reason I was asking is that I have some plans to build two/three output benchtop power supply using these modules. Using suitable transformer with three separate secondary windings is necessary as the current measurement in these modules is implemented as "low side current sense". My major concern has been how these modules behave on power on/power off and output enable/disable conditions. If I have understood correctly, the modules behave quite well and do not produce significant overshoot even with the larger load transitions. My intended use is mostly applications requiring few hundred milliamps ... 1 ampere with output voltage of 5V and thereabouts. Probably some output filtering is in order to reduce the ripple.

[StillTrying]

the current measurement in these modules is implemented as "low side current sense".

That's a bit of a nuisance.

[Kalvin]

the current measurement in these modules is implemented as "low side current sense".

That's a bit of a nuisance.

Definitely. High side current measurement would have allowed parallel modules (ie. multiple positive outputs) without galvanic isolation requiring individual secondary windings.

[Kalvin]

the current measurement in these modules is implemented as "low side current sense".

That's a bit of a nuisance.

However, it may be possible to hack the module to bypass the low side measurement and use an external high side current measurement circuit instead. I guess this is quite doable so one could build a power supply with multiple positive outputs using a transformer with a single secondary. If one wants to get a negative output as well, a transformer with two secondaries is needed.

[StillTrying]

Are we sure this dps 5005 uses "low side current sense". I can't find any reference to it needing at least one side's gnd isolated anywhere, and no one coming across a gnd loop problem.

[Kalvin]

Are we sure this 5005 uses "low side current sense". I can't find any reference to needing at least one side's gnd isolated anywhere, and no one coming across a gnd loop problem.

Pretty sure as measuring from the terminal OUT- to the current sense resistor has resistance less than 0.2 ohms.

[Kalvin]

Measuring resistance between the terminals IN- and OUT- shows more than 1 Mohm resistance when the module is powered off. It looks to me that there may be more problems when connecting the modules in parallel. My guess is that the output enable/disable is done in the ground side of the supply - however I cannot confirm that without reverse engineering the module. If that is a case, it looks to me that the best way of creating a power supply with multiple outputs is to use a transformer with multiple secondaries which will isolate the modules from each other.

Ps. I have DPS3005 modules, but the basic topology should be identical with the DPS5005.

[StillTrying]

"Measuring resistance between the terminals IN- and OUT- shows more than 1 Mohm resistance when the module is powered off."

LOL
Are +IN and +OUT still connected together when it's off.

Perhaps there's only you and me that thinks (knows!) this is a problem.
http://budgetlightforum.com/node/52853

[Kalvin]

"Measuring resistance between the terminals IN- and OUT- shows more than 1 Mohm resistance when the module is powered off."

LOL
Are +IN and +OUT still connected together when it's off.

Perhaps there's only you and me that thinks (knows!) this is a problem.
http://budgetlightforum.com/node/52853

Resistance between IN+ and OUT+ is high ohmic and capacitive - the DMM reading starts from around 500 kohm and goes downward, and switching the DMM probe polarity gives also quite similar results starting from 3 Mohm and going downward.

It looks to me that connecting these modules for multiple outputs without galvanic isolation (ie. a transformer with separate secondary windings) between the modules may be a bit problematic. Probably the module manufacturer could provide some detailed information.