Banggood PSU Enhancements

[shobo]

apparently a new version of the psu has appeared.i'll reply with the parts

[Cliff Matthews]

More info on new rev. http://www.icstation.com/product_document/Download/12479_installation_instructions.pdf

[soligen]

I didn't trace through this in detail, but at a glance it looks like the parts in the schematic have just been re-arranged in the drawing, and they drew in the 7824 regulator which was part of the kit I bought, just not on the old schematic Maybe some small changes - I didn't look at it in detail. The board does look like it has been re-laid out, but at a glance I don't see that the big flaws were corrected.

Are they still rating it 30V @3A?  If so still way to high.  My mods de-rated the design to 15V 2A

[Audioguru]

The "new" kit has the same defective circuit as the old Greek kit and later Chinese copies of it.
It will not produce regulated 30V at 3A and many of its parts are overloaded.

It still uses a 24VAC/3A transformer that will smoke and burn.
It still uses TL081 opamps that will also smoke and burn.
It is still missing voltage and current calibration trimpots.

[shobo]

yay onther one:
https://www.banggood.com/0-28V-0_01-2A-Adjustable-DC-Regulated-Power-Supply-DIY-Kit-Short-Circuit-Current-Limiting-Protection-p-1060253.html?rmmds=search&cur_warehouse=CN

This one is lower rated but looks simpler in its design with no op-amps.Perhaps all it's done by the atmega?

[Cliff Matthews]

So will you trust in current limiting from an MCU?
- Coded by people who's first line in an assembly manual says: "The first step is the Resistance Welding".
- And on first boot up the MCU displays: PSU by: https://shop35395693.taobao.com/
- https://item.taobao.com/item.htm?spm=a1z10.1-c.w4004-4669527207.21.12cf9858jeH4Gs&id=532555910832

Time for a smoke.. and I don't even smoke!. (Well, maybe a magic smoke... )

FWIW, Dave seems to be highlighting RDtech power modules recently and they seem to support him responsibly..
The newest DPH5005 Buck-Boost unit is output rated for 0-50v / 0-5a and max 250w for $34 (with the stock heat sink and 6-50v DC input, you could maybe trust it up to 100 watts perhaps..) just throw it into a case. WMMV

[shobo]

well after much thought i'll buy a pre-made one.i was eyeing an old romanian source like this:

https://marebazar.files.wordpress.com/2008/05/sursa.jpg

I would have prefered to get one of those kits since there available close by (the same city) and not wait for post and stuff but since i dont have a scope to trouble shoot it i guess buying something already made would be wise.

[MrAl]

Hello,

The original point of the thread was can a power supply be upgraded to allow for a higher output current.

The answer is yes but there is some serious changes that have to be made.

First, the output power transistors all have to be doubled if the current is doubled.
Second the input transformer current rating has to be doubled, the input diode bridge current rating hs to be doubled.
Third, the filtering capacitance has to be doubled.
Fourth, any current limit detection circuit has to be modified to allow the higher output current.
Fifth, the heatsinks have to be doubled or enlarged.
Last, it has to be tested at full load.

If it's an old design it should be upgraded with a reference diode rather than a zener which the old supplies used to use.  The reference diode is much more stable over temperature.

As a side note, you can build 5 amp power supplies with the higher current version of an LM317 and they are pretty simple with much less parts.  That will get you there pretty quick if you just want t linear.  IF you want a switcher it takes a little more work.

[Kleinstein]

The original circuit already needs modifications to make it work reliable at all. Likely higher voltage OPs - maybe limiting of the supply of there OPs that don't need to be so high. It's also a good idea to have something to prevent a voltage spike on turn on. As the current limit is rather slow, it might be a good idea to add a fixed fast current limit.

Increasing the power is only what comes after that.

[shobo]

well if i get one,i'll use it with a 18 v transformer to get a min of 24v

[tpowell1830]

From looking at the picture from Bangood, the 2 adjustment pots are too close together to be able to put very big knobs on them.

[Cliff Matthews]

TIP142's (TO-247) sell at just $1.50, it would be interesting to see how stable this would be, and where it needs help.
https://www.digikey.com/product-detail/en/stmicroelectronics/TIP142/497-2541-5-ND/603566

[shobo]

From looking at the picture from Bangood, the 2 adjustment pots are too close together to be able to put very big knobs on them.

They provide 2 cables with jst connectors to move the pots off board.

[Cliff Matthews]

They provide 2 cables with jst connectors to move the pots off board.

Yes, but Soligen's premise was to take this cheap $5 kit and improve it, so budget tech's will have a resource that's worthy of making this design into a respectable linear supply. I don't see any thread achieving that goal for the benefit of those not ready (or able) the shell-out big $'s. We just see post's about mods (here and on the net) while simultaneously getting shot down and being told to cop-out and buy somebody's pre-made xyz-3003 box. Poppy-cock.. That's NOT the true spirit of electronics.

[Kleinstein]

There is one main point to change with the old Banggood kit, and that is rather simple: use a lower voltage transformer. To stay inside the specified max. 36 V supply for an TL081 this would be about 18 V max, if the transformer is not that small so that the no load voltage will not be that much higher than nominal.  Thus would limit the maximum output voltage to about 18-20 V with light load and maybe 13-15 V under heavy load. At the reduced voltage the 3 A current rating might not be that bad, but a lower value (increase R18) might be better.

A bodged in diode from the negative output side to the base of Q1 might be a good idea too (this gives an additional fast current limit at about 4 A). Another point are decoupling caps at the OPs, especially the one for the output.

It is a little odd they have a 7824 regulator an do not use it for U1 and U3, this way they could have gone away with the TL081 for these two OP.

There are a few additional changes that would help, but than it gets difficult to keep the board.

The other option would be to change the OPs to something like MC34071 (instead of TL081), with a higher voltage rating. As this one is single supply one could also change D7 to a lower voltage one (e.g. 2 V) and Adjust R14 accordingly.

[soligen]

The 24V regulator was just added on to run 24V fan for the heat sink (not included).  Seems odd to me since 12V fans seem much more available. I used a SMPS module to run the fan at about 9V and did not use the 24V regulator.

For those who may not have read the whole thread, the original goal of increasing the specs on the kit was abandoned, and instead the specs were reduced to about 15.5 V max and 2.25A. (as well as some other mods too).  An 18V transformer was used, and I also needed to add significantly more input capacitance.

I had to increase output capacitance to get rid of some oscillations, so it also seems the design is also not necessarily stable without some tweaking.

I did finally get it working well, but it was a journey.  At least I learned stuff. 

[shobo]

The 24V regulator was just added on to run 24V fan for the heat sink (not included).  Seems odd to me since 12V fans seem much more available. I used a SMPS module to run the fan at about 9V and did not use the 24V regulator.

For those who may not have read the whole thread, the original goal of increasing the specs on the kit was abandoned, and instead the specs were reduced to about 15.5 V max and 2.25A. (as well as some other mods too).  An 18V transformer was used, and I also needed to add significantly more input capacitance.

I had to increase output capacitance to get rid of some oscillations, so it also seems the design is also not necessarily stable without some tweaking.

I did finally get it working well, but it was a journey.  At least I learned stuff.

Do an instructable of the mods

[shobo]

TIP142's (TO-247) sell at just $1.50, it would be interesting to see how stable this would be, and where it needs help.
https://www.digikey.com/product-detail/en/stmicroelectronics/TIP142/497-2541-5-ND/603566

Can you give me link to the tip142 supply?

[Cliff Matthews]

http://www.ve2ums.ca/chasse/Serge/Atelier/Projets/Membres/VE2EMM/alimentation_ang.htm
Scroll to the bottom for complete archive download (several versions inside). For further comments on the LM723 supply, I'd recommend opening another thread.
 
It appears Soligen has not given-up on the thread (and.. maybe I should delete my distracting posts?).

[soligen]

It appears Soligen has not given-up on the thread (and.. maybe I should delete my distracting posts?).

For me, this project is completed, so I don't care that you have added a different design to the conversation.  I am happy to have you leave your posts so anyone finding this thread has a reference to an alternative.

Although I got mine working well, I would not recommend this kit to anyone, so having an alternative listed is a good thing.

[shobo]

Found a modified version of the kit made by a guy in his garage but the differences are the resistors are half a watt and uses LM741 op-amps and at the same price as the kit.

https://img01-olxro.akamaized.net/img-olxro/175166427_4_1000x700_kit-modul-sursa-de-laborator-sursa-de-curentelectronic-arduino-electronice-si-electrocasnice_rev001.jpg

[jaycee]

You could do a lot worse than base your design around the ELV 22532 circuit. This is pretty much what the Mastech supplies are.

Some people dont like the idea of having to have a separate transformer for the bias supplies, but why not ? It only needs to be small. Mine uses a 9-0-9VAC transformer from an old digital clock/bedside radio... probably 2VA at the most, pretty small. It's enough to drive the opamps, a relay for changing between transformer taps, and run the display meters

[Cliff Matthews]

That's using LM741 a/b (44v) but not c version (same 36v problem as TL081). You pic comes from here I think..
https://www.olx.ro/oferta/kit-modul-sursa-de-laborator-sursa-de-curent-electronic-arduino-ID9v3vx.html
The highest voltage he shows is 28.5 and that's not even loaded..

I searched for schematic "sursa reglabila 0-30" it's the same. Maybe try this simple one with another 2n3055??
http://www.circuiteelectronice.ro/power-circuits/sursa-de-alimentare-reglabila-0-30v-cu-lm723.html

[shobo]

That's using LM741 a/b (44v) but not c version (same 36v problem as TL081). You pic comes from here I think..
https://www.olx.ro/oferta/kit-modul-sursa-de-laborator-sursa-de-curent-electronic-arduino-ID9v3vx.html
The highest voltage he shows is 28.5 and that's not even loaded..

I searched for schematic "sursa reglabila 0-30" it's the same. Maybe try this simple one with another 2n3055??
http://www.circuiteelectronice.ro/power-circuits/sursa-de-alimentare-reglabila-0-30v-cu-lm723.html

As an option i have this too. The guy sells only the pcb but not a kit.i might end up saying "fuck it" and buy one.

Trimis de pe al meu VFD 1100 folosind Tapatalk

[wrongtwins]

Hi , I have build the power supply but i dont have any voltage at the output ? Do you have any trouble shooting guide to help me out. I'm a hobbyiste i a dont know were to start to find the probleme. The 24V for the fan is working. At D1047 i have +28V, 0V, -28V.