Author Topic: DIY SCPI programmable dual channel bench PSU 0-50V/3A (now EEZ H24005)  (Read 451877 times)

0 Members and 2 Guests are viewing this topic.

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #200 on: November 15, 2015, 11:16:47 am »
Is this part of homework? I see too many formal maneurisms :D

I didn't get it. Please clarify.

Offline timofonic

  • Frequent Contributor
  • **
  • Posts: 904
  • Country: es
  • Eternal Wannabe Geek
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #201 on: November 15, 2015, 11:35:29 am »
Is this part of homework? I see too many formal maneurisms :D

I didn't get it. Please clarify.
Is this project part of homework for a subject of your studies?

 I understand some people don't like to show their "dirty work", but the common way in FOSS is "release early, release often" and there's no need to be ashamed of improving iteratively your stuff.

I can understand it is you require some football reviewing process because it's a project for your learning center.

Anyway, just an appointment. I was too curious, sorry for your annoyances.

Great work!
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #202 on: November 15, 2015, 01:26:58 pm »
Is this part of homework? I see too many formal maneurisms :D

I didn't get it. Please clarify.
Is this project part of homework for a subject of your studies?

 I understand some people don't like to show their "dirty work", but the common way in FOSS is "release early, release often" and there's no need to be ashamed of improving iteratively your stuff.

I can understand it is you require some football reviewing process because it's a project for your learning center.

Anyway, just an appointment. I was too curious, sorry for your annoyances.

Great work!

Thanks for clarification. I'm now closer to understanding :-+. This is my homework only in a way that I'm doing it from my home :). The new web pages that I started to build dedicated to this project will contain current status but also to some extent brief overview of past challenges, "dead-ends" and mistakes. In that way I think it could be more beneficial for people that don't want to goes thru all posts in this and few other threads, and also people who eventually follows mentioned posts don't want to read many now obsolete and irrelevant things once again.

Also there is no learning center. As it is said on the home page: all presented projects are created in a way to be suitable for DIY. Idea is to promote open source and contribute to the open source community. I think that chosen approach is not completely strange and unusual: creating hardware first as a reference design for the software that has a potential to be used with similar devices or even in the long run with the different type of devices within the same category (that means instruments like wave/function generators, DMMs, electronic loads, scopes etc.).

Your and other members input is welcomed, and hopefully it will follow and helps us to create something that makes sense. Many thanks!

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
New enclosure is arrived ...
« Reply #203 on: December 04, 2015, 08:49:46 am »
A new enclosure is arrived that is 40mm wider then previous one presented here. That one is huge enough to house larger transformer and that AUX power supply PCB is no more mounted on the top cover what is not so practical. The unpacked enclosure looks like this:



AUX power supply PCB is now mounted on the side support plate:



The front panel without final acrylic top mask looks like this:



The binding posts were extended using the 8mm spacer (marked red). The binding post's metal top was shorted for about 9mm before spacers can be properly mounted.



The USB and Ethernet sockets are now mounted on the rear panel:





When Front and rear panel is mounted with two side supports we got the following "frame" that can be easily separated from the bottom plate where main transformer is mounted (only one channel is shown):



I made a support for the main transformer because it was not possible to simply mount it without deforming a bottom plate:



... and with transformer mounted:



Previously mentioned "frame" with bottom plate that carry transformer now looks like this:





Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Completed enclosure
« Reply #204 on: December 04, 2015, 09:14:05 am »
I forgot to add a picture of the completed enclosure (with top cover mounted). Actually now only a front mask is missing.
Please also note that handles are not included in the enclosure package.




Offline timofonic

  • Frequent Contributor
  • **
  • Posts: 904
  • Country: es
  • Eternal Wannabe Geek
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #205 on: December 04, 2015, 12:26:05 pm »
I see you got anI already defined enclosure for a good price. That's nice!

I'm interested in getting enclosures for my future projects. I'm not sure if I can use generic ones or would need to use custom enclosures.

 How much did cost to you? I'm worried about the shipping costs.
« Last Edit: December 04, 2015, 12:32:47 pm by timofonic »
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #206 on: December 04, 2015, 08:05:07 pm »
I see you got anI already defined enclosure for a good price. That's nice!

I'm interested in getting enclosures for my future projects. I'm not sure if I can use generic ones or would need to use custom enclosures.

 How much did cost to you? I'm worried about the shipping costs.

Hi timofonic, this enclosure is not customized in any way for me.  All holes and cuts I made by myself with very basic and not precise tools, but at the end it looks fine. Cover mask that is missing on the front panel will hide TFT display border areas and make it even more visually appealing.

They are using only DHL and for Croatia it was 20EUR. They have online shipping calculator so you can check what does it mean for your destination.

Offline timofonic

  • Frequent Contributor
  • **
  • Posts: 904
  • Country: es
  • Eternal Wannabe Geek
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #207 on: December 08, 2015, 12:29:43 am »
I see you got anI already defined enclosure for a good price. That's nice!

I'm interested in getting enclosures for my future projects. I'm not sure if I can use generic ones or would need to use custom enclosures.

 How much did cost to you? I'm worried about the shipping costs.

Hi timofonic, this enclosure is not customized in any way for me.  All holes and cuts I made by myself with very basic and not precise tools, but at the end it looks fine. Cover mask that is missing on the front panel will hide TFT display border areas and make it even more visually appealing.

They are using only DHL and for Croatia it was 20EUR. They have online shipping calculator so you can check what does it mean for your destination.

That's interesting. Does the shipping costs get nearly the same if ordering a few of them?

I found the following projects, maybe they are somewhat similar. c4757p (aka Chris Palvina) collaborated in the project too, he's planning to do a FLOSS modular oscilloscope. He collaborated on other interesting Open Hardware projects

https://github.com/semianalog/mod12003
https://github.com/semianalog/mod1base

( https://github.com/WCP52/ )

I found this SCPI parser. Did you know about it?

https://github.com/j123b567/scpi-parser

(I asked both sigrok and scpi-parser projects about some collaboration, too)
 

Offline fuzzoli

  • Regular Contributor
  • *
  • Posts: 70
  • Country: us
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #208 on: December 08, 2015, 05:04:07 am »
Would it be possible to rotate the boards 90o?  It seems Eagle doesn't like the 60w x 100h dimensions.  The reason I'm asking is because I'd like to line up mounting holes on the Pre and Post Regulator boards so they could be mounted back-to-back.  Thanks!
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #209 on: December 08, 2015, 09:11:08 am »
That's interesting. Does the shipping costs get nearly the same if ordering a few of them?

I can confirm that at least for the quantity of 2. You can contact that guys they are very responsive and ask what max. weight is covered by stated shipping cost.

I found the following projects, maybe they are somewhat similar. c4757p (aka Chris Palvina) collaborated in the project too, he's planning to do a FLOSS modular oscilloscope. He collaborated on other interesting Open Hardware projects

https://github.com/semianalog/mod12003
https://github.com/semianalog/mod1base

( https://github.com/WCP52/ )

Thanks for this! I added semianalog into our list of related links (if someone knows for other that is not listed please let me know). It looks like a fully featured module that include MCU controller SMPS pre-regulator and dedicated OVP circuit. I'm wondering what is a "front end" for it: a PC or another dedicated MCU board that could handle more power modules.

I found this SCPI parser. Did you know about it?
https://github.com/j123b567/scpi-parser

Yes, of course :) as I stated that here. Please note that we are currently using our fork that includes some optimization for the AVR MCUs. Please also note that one can shortly found that MCUs with 32K of flash memory that are mentioned too often in DIY programmable power supply projects are far too small for decent coverage of PSU functionality of commercial middle of high end solutions (that almost without exception offers SCPI remote control). We currently can live with 256KB or 512KB MCUs but it's now much easier for us to imagine how to use something like BeagleBone if we'd like to stay completely into open hardware/open software arena or if not then solutions such as RasPi.

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #210 on: December 08, 2015, 09:14:53 am »
Would it be possible to rotate the boards 90o?  It seems Eagle doesn't like the 60w x 100h dimensions.  The reason I'm asking is because I'd like to line up mounting holes on the Pre and Post Regulator boards so they could be mounted back-to-back.  Thanks!

I'm not so sure what that mean that Eagle doesn't like mentioned dimensions? All PCB designs are now available here but please note that some of the components are locked, and you need to unlock them before rotation.

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Panelized PCB
« Reply #211 on: January 01, 2016, 06:10:58 pm »
Some people asked me for the panelized PCB for the complete solution (dual channel digitally controlled). You can find it now here: https://github.com/eez-open/psu-hw/tree/master/Panelized

Also a OSH Park profile with gerbers for the individual boards is also created, please visit https://oshpark.com/profiles/eez-open

Offline funshine

  • Newbie
  • Posts: 3
  • Country: cn
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #212 on: January 13, 2016, 05:25:55 am »
Hi,
Great job! I've read all the posts, and it's an amazing project.
I have a question related to mosfet pre-regulator, because I was planning to do a psu use the pre-regulator circuit.
from the site, http://www.envox.hr/eez/bench-power-supply/psu-pre-regulator.html, it's said
Status Completed, not recommended for toroidal main transformer.
in your post before, there is some big inrush current if a toroidal main transformer is used.
and also there is a soft-start circuit in main 220AC, here: http://www.envox.hr/eez/bench-power-supply/psu-auxiliary-supply.html.
I was wandering if this big inrush current is the only problem?
is there any other drawback for the mosfet pre-regulator used with toroidal main transformer?
Thanks!

I tried to add a soft-start to the pre-regulator, by monitoring the mosfet current, using the mosfet Rds-on as the sense resistor.
when the current is big enough, it turn on the SCR, and like the other part, turn on until next cycle.
attachment is the LTspice and some screen shot.
no real test in breadboard or PCB, maybe it can solve the inrush problem?
 

Offline funshine

  • Newbie
  • Posts: 3
  • Country: cn
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #213 on: January 13, 2016, 05:27:54 am »
sorry missed the screen shot
 

Offline skrap

  • Contributor
  • Posts: 29
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #214 on: January 13, 2016, 05:41:11 am »
Nice work! I really like it :-+

Any particular reason you picked the venerable TL07x in the post reg.?
I like the E36xx-ish style but a TL07x replacement like for example the TI suggested TLE2072 or TL052 would have been a nice improvement.

Why did you decide on a discrete current mirror since MMBT5401 is a single package and won't matched easily?
I guess you already know about the typical modern alternatives (Fig.5 or Fig.12  http://www.analog.com/library/analogdialogue/archives/48-04/current_output.html ).

Just a side note; I don't know if you noticed this when looking at the E36xx and similar HP bench PSUs but one of my favorite tricks they used is to let the opamps reference the output as GND so that V+ and V- is always centered around it. That's how they get a way with high-side current sensing for example and it also simplifies the loop-design itself quite a bit.
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #215 on: January 13, 2016, 08:51:12 am »
Hi,
Great job! I've read all the posts, and it's an amazing project.
I have a question related to mosfet pre-regulator, because I was planning to do a psu use the pre-regulator circuit.
from the site, http://www.envox.hr/eez/bench-power-supply/psu-pre-regulator.html, it's said
Status Completed, not recommended for toroidal main transformer.
in your post before, there is some big inrush current if a toroidal main transformer is used.
and also there is a soft-start circuit in main 220AC, here: http://www.envox.hr/eez/bench-power-supply/psu-auxiliary-supply.html.
I was wandering if this big inrush current is the only problem?
is there any other drawback for the mosfet pre-regulator used with toroidal main transformer?
Thanks!

I tried to add a soft-start to the pre-regulator, by monitoring the mosfet current, using the mosfet Rds-on as the sense resistor.
when the current is big enough, it turn on the SCR, and like the other part, turn on until next cycle.
attachment is the LTspice and some screen shot.
no real test in breadboard or PCB, maybe it can solve the inrush problem?

Thanks funshine for your comments. Notice/disclaimer about pre-regulator is valid, and I didn't have a chance yet to elaborate that on this forum but now it's a time. The issue with such kind of pre-regulator is huge current peaks that resulting in excessive heating of main transformer. That is much more obvious in case of toroidal transformer for two reasons: it's primary and secondary windings are more hardly coupled, and core is inside (covered with windings) what make its cooling more difficult. I realized that in latest stage when everything was completed. As you probably know that is not my design, nor I tried to make any chances to eventually improve it. There was no warning that one should at least avoid toroidal type of transformer or at least I didn't remember that blackdog who presented this circuit here mentioned anything like that. Unfortunately, I missed opportunity to ask a right question. And yes, as it's usually a case right questions usually moves us in right direction but asking right question require right level of experience and knowledge :).
BTW, blackdog presented this pre-regulator on the thread with right title started by Kalvin: Very Low Noise Preregulator for Benchtop Power Supply. I didn't ask if he eventually find something or not. There is another attempt started by tombi (Tom) based on the same "blackdog"'s pre-regulator (put into quotation since blackdog said that he is not author). Tom's progress could be follow here, and I warned him about possible issue with main transformer (over)heating. On another (not English language) forum was discussion about pre-regulator and I can say some real experts told me such behavior is expected and that Agilent uses special designed transformers when is using SCR or mosfet phase controlled pre-regulation. Interesting thing is also that people on that forum find few errors in schematic of Agilent model E3640 where pre-regulator with SG3525 PWM controller! We don't know if that errors are intentional or not but PWM controller cannot work with presented values. Some of the member from the forum succeed to finally run it and answer to my question how hot is transformer was unfortunately the same: its start to be hot pretty soon with higher currents.
To summarize I'd like to say what is possible with current solution: on the transformer I can measure 48oC when current is 2 A. But, with the 5 A SMPS pre-regulator (shown below) temperature rises to 46oC on the same transformer with 4 A! Please take a note that no forced cooling is deployed, and I didn't tested such combination since I'd like to stick with initial goal to not use (noisy) fans what is btw a standard.



For higher currents sourced on both channel transformer temperature could easily rise over 70oC. Not nice at all, so NTC is added to transformer to avoid possibly catastrophic results. So, in general this solution WITH toroidal transformer and WITHOUT any forced cooling is pretty usable for output currents of up to 2 A.

Regarding added inrush current limiter and soft-switch: it was not added to resolve issue with continuous work of pre-regulator but only address initial in-rush current and serves as convenient MCU driven main switch.

It's interesting that so many attention and energy is spent of power supplies without pre-regulation (actually that is post-regulator or final stage in our design) and many people is possibly not aware of the problem when someone presented e.g. linear 0-30 V / 2 A without disclaimer (or awareness) or that it cannot deliver 2 A on 3.3 or 5 V continuously since that requires ~50 W of dissipation on heatsink which is too small or that is sometimes outside pass-element's SOA. From other side many people believes that SMPS (e.g. buck) pre-regulation could solve that problem but didn't think or want to talk about huge output noise that easily bypass linear post-regulator and is still present on the output. Again I'm talking about currents of 2 or more amps. The latest and (first?) resort is main transformer with multiple secondary windings that are switched on sequentially in old-fashioned way with power relays or in some more advanced designs with SCRs or mosfets. If multiple windings transformer is acceptable solution then I'd rather like to recommend Liv's design that is superior to mentioned relays, SCRs or mosfets but is not so simple.

The story about pre-regulator and complete power supply is not finished yet. It's a work in progress but now I can say for the first time that I have a real platform that can be used on daily basis without fear that it will kill the connected load/device due to some error in hardware and/or firmware. A real achievement since our primary "expertise" is software not hardware and we learned a lot about different aspect of building and hopefully all that is presented could help other people to make even better solution that is O P E N and suitable for DIY.

Stay tuned ... and I'd like to thanks in advance for all your further inputs.

Offline Liv

  • Regular Contributor
  • *
  • Posts: 134
  • Country: by
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #216 on: January 13, 2016, 09:01:37 am »
it's said Status Completed, not recommended for toroidal main transformer.

If you have in mind when the pre-regulator working at the mains frequency, using a toroidal transformer will be a big peak current when the MOSFET is turned on because of the low leakage inductance. For example, the Agilent U8002 to do two things: in series with the secondary winding of a toroidal transformer added bulk choke and MOSFET switching speed is limited. Heat slightly increased, but due to the low operating frequency efficiency falls not much.

Why did you decide on a discrete current mirror since MMBT5401 is a single package and won't matched easily?

This scheme is not required for high accuracy current mirror, so the cheapest solution suited to discrete transistors.

my favorite tricks they used is to let the opamps reference the output as GND so that V+ and V- is always centered around it. That's how they get a way with high-side current sensing for example and it also simplifies the loop-design itself quite a bit.

Not a bad topology. But then on the output voltage will be the processor, display, control buttons. For the amateur construction this may be unsafe. Agilent usually divides the processor into two parts, one of them floating (output referenced), other earth referenced. They are connected via optocouplers. But this complication of the structure.
« Last Edit: January 13, 2016, 09:10:36 am by Liv »
PSU PSL-3604 Pulse gen. PG-872 Freq. cnt. FC-510
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #217 on: January 13, 2016, 09:06:44 am »
Nice work! I really like it :-+

Any particular reason you picked the venerable TL07x in the post reg.?
I like the E36xx-ish style but a TL07x replacement like for example the TI suggested TLE2072 or TL052 would have been a nice improvement.

Hi skrap, dual op-amp standard SO-8 packaging are intentionally used on the PCB just to allows one to experiment with different type of op-amps! Yes, mentioned TLE2072 (or maybe TLE2022) looks as good alternative, and someone has to test it.

Why did you decide on a discrete current mirror since MMBT5401 is a single package and won't matched easily?
I guess you already know about the typical modern alternatives (Fig.5 or Fig.12  http://www.analog.com/library/analogdialogue/archives/48-04/current_output.html ).

See Liv's answer

Just a side note; I don't know if you noticed this when looking at the E36xx and similar HP bench PSUs but one of my favorite tricks they used is to let the opamps reference the output as GND so that V+ and V- is always centered around it. That's how they get a way with high-side current sensing for example and it also simplifies the loop-design itself quite a bit.

Yes, I did notice that. But such design requires also floating bias supply for op-amps and I tried to avoid complex multiple windings main transformer or adding a small PCB transformer on the pre-regulator PCB. The AC mains is present only on one PCB (AUX PS) that makes a whole design safer.


Offline skrap

  • Contributor
  • Posts: 29
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #218 on: January 13, 2016, 12:03:27 pm »
This scheme is not required for high accuracy current mirror, so the cheapest solution suited to discrete transistors.

What I meant was that the current mirror seems unnecessary complex if you aren't going to match the transistors. For that a unipolar current source would work just fine instead, one transistor and a few resistors would suffice, similar to Q9 in the circuit for example.

Yes, I did notice that. But such design requires also floating bias supply for op-amps and I tried to avoid complex multiple windings main transformer or adding a small PCB transformer on the pre-regulator PCB. The AC mains is present only on one PCB (AUX PS) that makes a whole design safer.

Ah I see, I never thought of it that way. I've always separated the digital and analog supplies due to noise requirements but those were not 0-50V/3A supplies ;-)
 

Offline Liv

  • Regular Contributor
  • *
  • Posts: 134
  • Country: by
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #219 on: January 13, 2016, 12:31:34 pm »
What I meant was that the current mirror seems unnecessary complex if you aren't going to match the transistors. For that a unipolar current source would work just fine instead, one transistor and a few resistors would suffice, similar to Q9 in the circuit for example.

In this case, need on/off current source, the on/off signal must be TTL referenced to ground. It can be a single transistor current source, but the on/off circuit will give the same complexity as the current mirror.

In my PSL-3604 used several current mirrors. This has the advantage of being easy to switch off all together, and also to set the operating current by a single resistor. The accuracy there is also not required, which allows to implement current mirrors with discrete transistors.
PSU PSL-3604 Pulse gen. PG-872 Freq. cnt. FC-510
 

Offline skrap

  • Contributor
  • Posts: 29
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #220 on: January 13, 2016, 11:59:24 pm »
In this case, need on/off current source, the on/off signal must be TTL referenced to ground. It can be a single transistor current source, but the on/off circuit will give the same complexity as the current mirror.

You can simplify it by removing Q7 and replacing it with a resistor instead and then just tweak the value of R10 for the desired bias. Using a transistor for Q7 really makes the most sense when you create circuits on a wafer where resistors is not only harder to achieve, especially on waferscale, but you also get additional transistors essentially for free.

But if you prefer a transistor over a resistor then why not. It's not like they are expensive anyway :-)

Bellow is a crude example of replacing the transistor and still keep the Q10 on/off-switch and have it connected to the Q11 master-switch. One could even use Q11 directly and skip Q10 and have it control several current sources like you described for PSL-3604. Nice project btw!
 

Offline funshine

  • Newbie
  • Posts: 3
  • Country: cn
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #221 on: January 14, 2016, 02:45:07 am »
Thank you for the quick reply!
so the toroidal transformer get hot because the peak current, caused by rectifier voltage much higher than the bulk capacity when mosfet turn on/off.
I guess the lower the output voltage, the higher the transformer's temperature, if the output current stays the same.
so if I replace with R-core transformer, will it be better?

and one more question, if a mosfet pre-regulator used, can the output current get higher when output voltage is low?
for example, a 50VA transformer, ideally 0~25V/ 2A output when no pre-regulator used, can serve only 2A when output voltage is 5V.
with the mosfet pre-regulator, can it serve 4A when 5V output?

I was care about the switching noise, so if there is any chance, I will not add smps pre-regulator, but rather go for the multi-windings.
although with smps pre-regulator, if the switch rise/fall time is controlled very carefully, there maybe some noise?
but agilent E36xx series doing quiet well, don't know whether I can do the same noise level.

Thanks in advance!
 

Offline Liv

  • Regular Contributor
  • *
  • Posts: 134
  • Country: by
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #222 on: January 14, 2016, 09:24:14 am »
One could even use Q11 directly and skip Q10

Supply voltage for the current source is not stable. Using simple resistor divider we have output current variations. Common-base Q10 uses regulated +5 V supply, so the collector current (and output current of current source) will depend on the R27 value only. Q7, in principle, can be eliminated. But even with discrete transistors, the temperature stability of the current source would be better. How much - it depends on the ratio of the B-E voltage and the voltage drop across R11.

I was care about the switching noise, so if there is any chance, I will not add smps pre-regulator, but rather go for the multi-windings.

I have the same position, so applied to the PSU multi-level output stage.
PSU PSL-3604 Pulse gen. PG-872 Freq. cnt. FC-510
 

Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
Re: DIY programmable dual channel bench PSU 0-50V/3A
« Reply #223 on: January 14, 2016, 04:44:55 pm »
Thank you for the quick reply!
so the toroidal transformer get hot because the peak current, caused by rectifier voltage much higher than the bulk capacity when mosfet turn on/off.
I guess the lower the output voltage, the higher the transformer's temperature, if the output current stays the same.
so if I replace with R-core transformer, will it be better?

Sorry, I'm not competent to comment this. Hopefully some other member could explain you the theory. Someone need to test it with R-core or at least with EI-core transformers, neither of them with 100 or more VA is on my disposal.

and one more question, if a mosfet pre-regulator used, can the output current get higher when output voltage is low?
for example, a 50VA transformer, ideally 0~25V/ 2A output when no pre-regulator used, can serve only 2A when output voltage is 5V.
with the mosfet pre-regulator, can it serve 4A when 5V output?

Tested with single mosfet up to 7 A but I cannot remember what output voltage was set.

I was care about the switching noise, so if there is any chance, I will not add smps pre-regulator, but rather go for the multi-windings.
although with smps pre-regulator, if the switch rise/fall time is controlled very carefully, there maybe some noise?
but agilent E36xx series doing quiet well, don't know whether I can do the same noise level.

Thanks in advance!

If I remember correctly in E36xx is not using SMPS pre-regulation.
Maybe solution that is proposed some time ago by void_error could be satisfactory for many people: using smps with pmos switcher that can provide 100% duty cycle: that means that with MCU power control you can handle dissipation on post-regulator pass element when you need "silent" source (without HF switching noise). Here is an example:



Offline prasimixTopic starter

  • Supporter
  • ****
  • Posts: 2022
  • Country: hr
    • EEZ
LTC's AN104 load transient tester PCBs
« Reply #224 on: January 17, 2016, 01:22:15 pm »
A new GitHub repository is now available for LTC's AN104 closed loop load transient tester that is mentioned here and some people asked for it.
I create two PCB variants: with and without on board power supply section.

PCBs are also available on OSH Park here and here.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf