Author Topic: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)  (Read 451909 times)

0 Members and 3 Guests are viewing this topic.

Offline VEGETA

  • Super Contributor
  • ***
  • Posts: 1916
  • Country: jo
  • I am the cult of personality
    • Thundertronics
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #500 on: November 24, 2016, 12:11:29 am »
you said your supply is 0-3A so what if the load actually requires more than 3A? how can the design limit that without CC circuit being activated?

Offline void_error

  • Frequent Contributor
  • **
  • Posts: 673
  • Country: ro
  • I can transistor...
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #501 on: November 24, 2016, 12:29:49 am »
It goes into CC mode as soon as the current drawn reaches 3A or whatever it's set to below 3A. If your load requires more current you need a bigger power supply.
Trust me, I'm NOT an engineer.
 

Offline VEGETA

  • Super Contributor
  • ***
  • Posts: 1916
  • Country: jo
  • I am the cult of personality
    • Thundertronics
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #502 on: November 24, 2016, 12:46:25 am »
I understood the thing as follows: there is an already 3v on the comparator as a limit made by the designer, so the moment the supply is on, this set voltage is fed into the current limit comparator (assuming 3v = 3a). The the user may pick any value he wants if he wanted say like 2A. Is this correct? where can I see this in the circuit (the preset of the assumed 3v voltage)?

BTW, this current limiting circuit didn't work in LTSpice. Is this the case for you?

thanks

Offline void_error

  • Frequent Contributor
  • **
  • Posts: 673
  • Country: ro
  • I can transistor...
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #503 on: November 24, 2016, 01:12:59 am »
For the latest schematic the CC loop is built into IC35 and a DAC (IC39) is used to set the current limit. R33 is used by both IC35 and IC36 as the current sense resistor.
Trust me, I'm NOT an engineer.
 

Offline VEGETA

  • Super Contributor
  • ***
  • Posts: 1916
  • Country: jo
  • I am the cult of personality
    • Thundertronics
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #504 on: November 24, 2016, 01:19:08 am »
For the latest schematic the CC loop is built into IC35 and a DAC (IC39) is used to set the current limit. R33 is used by both IC35 and IC36 as the current sense resistor.

i mean does the dac outputs max limit by default from the start (by software) or what is the way?

Offline void_error

  • Frequent Contributor
  • **
  • Posts: 673
  • Country: ro
  • I can transistor...
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #505 on: November 24, 2016, 01:25:33 am »
i mean does the dac outputs max limit by default from the start (by software) or what is the way?
The DAC output voltage is controlled by software. I assume it outputs 0V at start-up but I haven't read the datasheet.
Trust me, I'm NOT an engineer.
 

Offline VEGETA

  • Super Contributor
  • ***
  • Posts: 1916
  • Country: jo
  • I am the cult of personality
    • Thundertronics
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #506 on: November 24, 2016, 01:50:49 am »
Well, if the design outputs 0v at startup and by default then it is not limited to 3A at all.

Offline void_error

  • Frequent Contributor
  • **
  • Posts: 673
  • Country: ro
  • I can transistor...
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #507 on: November 24, 2016, 04:00:58 am »
In fact it is. Take a look through the LT3741 datasheet. Output current is determined by the shunt resistor (R33 in this case) and the voltage at the CTRL1 or CTRL2 pins.

For this particular application prasimix chose the value for R33 for a maximum current of 3A. In that case the CTRL pin voltage will be 1.5V. According to the satasheet anything higher than 1.5V will not set the current any higher. If you would want to set the current limit to 1A you'd have to apply 0.5V to the CTRL pins, for 2A it would be 1V, etc.
Trust me, I'm NOT an engineer.
 

Offline VEGETA

  • Super Contributor
  • ***
  • Posts: 1916
  • Country: jo
  • I am the cult of personality
    • Thundertronics
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #508 on: November 24, 2016, 06:32:38 am »
I am seeing the consolidated pdf and I_Set is coming from this DAC: DAC8552IDGKTG4 which I don't see the said resistors so maybe it is just software.

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #509 on: November 24, 2016, 07:35:39 am »
Isn't the INA282 a bit overkill? I mean with 20uV offset it's really accurate but it's also quite pricey. I assume you wanted to share R33 from the LT3741.

Nice point void_error. It's a bit of overkill. It's not necessary that R33 is shared I just find it convenient not to introduce another shunt resistor. Hopefully it will works as is, I mean current monitor inputs shouldn't interfere with LT3741 inputs. In not I'll add another shunt resistor.
Do you have any other device on mind? Perhaps INA199, it's very cheap and TME have it.

Offline void_error

  • Frequent Contributor
  • **
  • Posts: 673
  • Country: ro
  • I can transistor...
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #510 on: November 24, 2016, 07:59:39 am »
Nice point void_error. It's a bit of overkill. It's not necessary that R33 is shared I just find it convenient not to introduce another shunt resistor. Hopefully it will works as is, I mean current monitor inputs shouldn't interfere with LT3741 inputs. In not I'll add another shunt resistor.
Do you have any other device on mind? Perhaps INA199, it's very cheap and TME have it.

I was also looking at a TI part. The INA199 seems to be ok, typical offset of 5uV and 150uV max.
As far as distributors go I found one for Farnell that also sells to individuals for my country (see flag) so I'm not limited to TME anymore. I'll drop the link here just in case anyone is interested. Haven't ordered anything from them yet.

Other than that I have a question: How did you figure out the down programmer? Got any tips?
Trust me, I'm NOT an engineer.
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #511 on: November 24, 2016, 08:04:45 am »
@VEGETA: Current limit indeed works in LTspice if you read datasheet and see what to change. See for example half-scale (2.5 A) simulation:



Please also read once again #498 to learn that we are talking here about new proposal: a 3rd channel that will accompany existing two power channel. Therefore referring to DAC8552 that resides on Power board is not applicable here. DAC what is proposed for this channel is MCP4822.
Also take note that R33 is set for max. current of 5 A not 3 A (check LT3741 datasheet pg.13).


Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #512 on: November 24, 2016, 08:10:48 am »
Other than that I have a question: How did you figure out the down programmer? Got any tips?

Actually cannot say anything meaningful. That part comes directly from Liv design.

Offline void_error

  • Frequent Contributor
  • **
  • Posts: 673
  • Country: ro
  • I can transistor...
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #513 on: November 24, 2016, 07:58:21 pm »
With a bit of help from a few forum members I managed to figure it out. The way Liv has drawn it it looks more like a class AB amplifier which can source a few times more current than it can sink and it also acts like one. It appears to be a very stable design and similar to what's used in a few HP/Agilent/Keysight/whatever power supplies. I managed to tweak it so it works with BJTs instead of MOSFETs because it's easier to parallel them for linear operation.

I also did some simulations on the circuit and they can be found here if anyone is curious.
Trust me, I'm NOT an engineer.
 

Offline VEGETA

  • Super Contributor
  • ***
  • Posts: 1916
  • Country: jo
  • I am the cult of personality
    • Thundertronics
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #514 on: November 24, 2016, 10:10:40 pm »
Ok, can you post a CV CC simulation in LTSpice that works?

Offline Jbates58

  • Regular Contributor
  • *
  • Posts: 68
  • Country: au
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #515 on: November 26, 2016, 12:19:36 pm »
hi all

just wondering if it was to make this PSU (as i am looking for a decent bench PSU for my next toy) i am just wondering what i can do in reguards to beefing up the current?

i do a fair bit of auto motive stuff and would love to be able to push up to 20ish amps. more would be even better. but i am unsure what to do to beef it up to do that.

cheers

Jason
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #516 on: November 26, 2016, 12:52:05 pm »
Current design allows MCU controlled paralleling of two channels to get 10 A output.



Beefing up could include the following:
  • Pre-regulator: Paralleling Q1 and decrease shunt resistor R2, change power inductor L2 with one that has higher Irms and Isat or paralleling two of them with higher inductance or with same inductance but higher switching frequency. L3 and L4 has to be changed to withstand higher current or you can remove that filtering section.
  • Post-regulator: Change current shunt resistor value (R69) and/or change current monitor (IC8) gain. It's quite possible that paralleling of Q4 will be also required or you can lower voltage range to e.g. 30 V to have a better SOA margin of the currently selected mosfet.
If you succeed to push both channel to 10 A with paralleling you will get 20 A. BTW, firmware support any voltage and current range, it's a question of changing few constants in the config file.

Offline KarolNi

  • Newbie
  • Posts: 6
  • Country: pl
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #517 on: November 30, 2016, 06:15:53 pm »
Hey mates!
I'm thinking about new bench supply. Do you know anything about crowdsourced buy, yet? Might be p'n'p or kit.
Cheers
KarolNi
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #518 on: November 30, 2016, 10:10:04 pm »
Thanks for asking. Everything is delayed because of weird situation with sending one unit to CS guys for preparing promotional video and sending it around for reviews. It reached their address three weeks ago, and US post send it back the same day because shipping address is supposedly not correct! And it's now who knows where. I have another two but I cannot send any of them because we need it for firmware development and testing. So we'll probably not continue as initially planned and launch campaign without any 3rd party review.

I'm planning few pledge levels: from set of bare PCBs to set of all preassembled modules and wire harness that has to be mounted into customized metal enclosure. No completed unit is planned since certification cost is insane (I got an offer for CE and FCC that is about 10.000 EUR!). That could make sense only if 500 or more units is shipped what is, I'm afraid, way beyond potential of crowdfunding platform since unit cost is not like 30 EUR but tenfold higher.

So, please be patient and stay tuned. In the meantime I'll report progress on firmware side and some thoughts about possible upgrade like e.g. recently announced 3rd channel with USB changing capability, etc.

Offline VEGETA

  • Super Contributor
  • ***
  • Posts: 1916
  • Country: jo
  • I am the cult of personality
    • Thundertronics
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #519 on: December 01, 2016, 12:59:07 am »
May I ask how much did it cost and what is the target price? how do you make it worth it while being somehow profitable for you?

thanks!

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #520 on: December 01, 2016, 09:17:59 am »
The campaign is conceived as a group buy and should promote free and open source hardware and software. When we comes to term "profitability" even if it for the most people means financial gain on the first place that is not the only possibility. I already profit from this project by learn a lot about different things and by getting a solution that is under my control and that can be almost infinitely modified on the firmware level in accordance with my past, present and future needs. The final price is not set yet, but it will be set to the point that it don't generate any financial loss to me. Therefore I have to take care to not undercalculate and if campaign exceed set target I'm going to add some extra stuff or in "worst" case start to donate extra "units" (modules/kits) primarily to educational institutions if someone will be interested.

The first campaign target is massive 100 units. I don't know if that is possible to reach but below that costs of manufacturing (BOM+PCBA, enclosure, wire harness, transportation box) becomes unattractive.

 
The following users thanked this post: vinicius.jlantunes

Offline vinicius.jlantunes

  • Regular Contributor
  • *
  • Posts: 225
  • Country: br
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #521 on: December 01, 2016, 11:00:00 am »
prasimix, I have quietly followed this thread and am eagerly waiting for the campaign to go-live. I live in a far away country with harsh import duties but do hope the price will be affordable by me and if so will definitely get in.

Either ways I wanted to say your work is admirable. Congrats and all the best in your campaign and any other endeavors!

Offline Theboel

  • Frequent Contributor
  • **
  • Posts: 278
  • Country: id
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #522 on: December 01, 2016, 12:42:51 pm »
Prasimix,

could You stop for a while made this beauty available to whoever want it and then listen to their feed back and start "B version"
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #523 on: December 09, 2016, 11:35:26 am »
I opened a new topic about crowdfunding related discussion since this forum has a separate section for that. Here is the link.

Offline alekso56

  • Newbie
  • Posts: 1
  • Country: no
Re: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)
« Reply #524 on: December 10, 2016, 08:54:19 pm »
Hi guys, im not sure where to post this, but i couldn't wait on the crowdfunding thing so i bought about 5 panels, and i have started diy'ing the project, so i have about 4 extra red panels cut with tabs...
Git revision: r5b9
 i can resell them here for about 33$ per complete panel (shipping not included). (i make zero profit off this, it's just to make my money back)

 
The following users thanked this post: prasimix


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf