DC Buck Step Down Converter DC 5A Regulator Voltage Current Adjustable DE

[ez24]

Hi

I bought some of these:

      "DC Buck Step Down Converter DC 5A Regulator Voltage Current Adjustable DE"

From :

http://www.ebay.com/itm/400714751657?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

(just a random picked seller)

Here are the specs and I have questions on what some of the specs mean:

==========================================================================
    Module Properties: non-isolated step-down module (BUCK)

-- what does non-isolated mean in this case ?  Is it the neg of the input is directly connected to the output ?
--  I will look up  "BUCK"

---------------------------------------------------------------------------
    Input voltage: 4-38V

--  ok  I will be using a 19 v  laptop power supply


---------------------------------------------------------------------------
    Output voltage:1.25-36V continuously adjustable

--  good this is what I want (I think)  SO  if I want  say  3.3, 5.0, 9.0, 12.0  volts - I would use 4 of these and set them to the correct voltage ?   And one for a variable range  ie 5 (bummer I ordered 4)


---------------------------------------------------------------------------

    Output current :0-5A Adjustable (recommended for use within 4.5A)

--  they advertize 5 amp but warn 4.5 a    BUT what do you think the real amps are based on the components

---------------------------------------------------------------------------
    Output power: recommended for use in below 75W, more than 50W please add heat sink. (Heat sink not included)

-- how to add a heat sink and I assume it would go on the chip that is laying flat on the circuit board

-- so using 4.5 amps and 50 w as limiting factors here is what I come up with

3.3 volts   at  4.5 amps  -->  P=IE  P = 14.9 watts  < 50 watts  so set I to 4.5 amps at 3.3 volts

5.0 volts  at 4.5 amps  -->   P=IE   P = 22.5 watts < 50 w  so set I to 4.5 amps at 5.0 volts

9.0 volts  at 4.5 amps  -->  P = 40.5   so set I to 4.5 amps at 9 volts

12 volts  at 4.5 amps  -->  P  = 54 watts  > 50 watts   close but I would de-rate this to 4 amps.  I think with this kind of module you absolutely cannot go over.  If fact if I wanted to use 4 amps I would do a lot of testing on this.  So I think I would rate this 3 amps at 12 volts

Am I correct in these calculations ?

I would probably change these numbers based on testing and temps of the components

---------------------------------------------------------------------------

    Working temperature: -40 to +85 degrees

-- ok

-----------------------------------------------------------------------------
    Operating Frequency: 180KHz

--  no idea what this means  on a DC circuit
---------------------------------------------------------------------------

    Conversion efficiency: up to 96%

--  ok

---------------------------------------------------------------------------
    Load regulation: S (I) ? 0.8%

--  I do not know what this means
---------------------------------------------------------------------------

    Voltage Regulation: S (u) ? 0.8%

--  I know what voltage regulation is but I do not know what  S (u) ? 0.8%  means
---------------------------------------------------------------------------

    Short-circuit protection: Yes (the limited current is 8A)

--  I know what short circuit protection means  but not "the limited current is 8A"   Does this mean no matter what happens it shuts down at 8 amps ?   ie when a short circuit gets to 8 amps it shuts off ?  AND anything else that makes it get to 8 amps

So if I had a 5 v circuit that started to draw 8 amps it would shut down?


---------------------------------------------------------------------------

    Over-temperature protection: Yes (automatic shutdown when over-temperature)

-- ok but what temp ?  could I test this by making it get hot and check the temps.

---------------------------------------------------------------------------

    Input reverse polarity protection: No, (if necessary, please string into the high current diode)

--  interesting  I think with my adapters this would not happen but how to add a high current diode ?  Sounds like an expensive component
---------------------------------------------------------------------------


    Installation: four 3mm screws

--  I assume  3 mm holes


---------------------------------------------------------------------------

-- How could I failure test these ?  At $4 each I think I could failure test 10 units.

--  Is there anyway to find a schematic?  Would I have to reverse engineer it to come up with a schematic ?

I would appreciate a testing protocol

thanks
Roger

[kjs]

Many questions and I will try to answer a few.
Non-isolated means that input ground and output ground are directly connected together.

There are buck converters (step down) and Boost converters (step up). Buck means that the output voltage always has to be lower than the input voltage. The device also needs some headroom (usually 2V0. This means that with a 19V input supply your maximum output voltage is ~17V. Please be also aware that these converters can only go down to ~1.25V output voltage.

The real limit is the maximum power and that's 50W (VERY optimistic) w/o heatsink. You can put a tiny heatsink on top but the better way to cool it is a heatsink on the non-component side of the PCB (and a LARGE one). The IC chip is mounted on the leadframe to get the heat to the PCB and there are via holes which transfer it to the backside. At 50W that thing gets pretty hot.

this is a switchmode power supply and the switcher runs at 180kHz. This also means that the output is pretty noisy and not what you would expect from a real lab power supply.

I haven't been able to read the IC name but it's something like the LM2596. Look at that datasheet and you know more even if it's one of the Chinese copies.

Edit: the 95% efficiency is the best you will ever see at one specific input/output voltage and current. Don't expect to have that where you want to operate.

[ez24]

thank you kjs

Non-isolated means that input ground and output ground are directly connected together. 

-- thanks

There are buck converters (step down) and Boost converters (step up). Buck means that the output voltage always has to be lower than the input voltage.

 

 -->  simple answer tks

The real limit is the maximum power and that's 50W (VERY optimistic) w/o heatsink. You can put a tiny heatsink on top but the better way to cool it is a heatsink on the non-component side of the PCB (and a LARGE one). The IC chip is mounted on the leadframe to get the heat to the PCB and there are via holes which transfer it to the backside. At 50W that thing gets pretty hot.

I spotted a module with a tiny heat sink on the LM2596 and a slightly larger one on the back.  I never saw a heat sink on a pcb so I thought it was a joke and now I cannot find it.  I will keep looking for it.  And if I remember the current max was 2 amps with a heat sink so now  I think that was real

I do not believe 50w either so I got a couple of these:


[url=http://www.ebay.com/itm/300W-DC-DC-CC-CV-Buck-Converter-7-40V-to-1-2-35V-8A-Step-down-Power-Module-/261530423651?pt=LH_DefaultDomain_0&hash=item3ce46d9163]http://www.ebay.com/itm/300W-DC-DC-CC-CV-Buck-Converter-7-40V-to-1-2-35V-8A-Step-down-Power-Module-/261530423651?pt=LH_DefaultDomain_0&hash=item3ce46d9163[/url]

At first I thought this was 2 linear chips on it but I could not understand the pin layout on the back.  After watching some Youtube
videos and learning a little more about these things,  I think this has 1 linear (for 5 volt) and probably a LM2596.   There was one video where he took off the components (a different module) and there was a LM2596, a 5 v regulator for a dual op amp and a transistor.  I think this is what is on these modules I have bought.

I will be using these modules to max out a laptop power supply. 

I have to set all the modules current total to what the laptop supply can handle.  I think I will have to use a spreadsheet.

this is a switchmode power supply and the switcher runs at 180kHz. This also means that the output is pretty noisy and not what you would expect from a real lab power supply.

Thanks  -  I will work on the noise as part of my learning experience.  I ordered a couple of used Power Designs power supplies to compare with  (looking for a 5005 - love at first sight).  I also bought a used HP1200A scope because it has a vert/div of 100 microvolts.  But the CRT broke loose and I am waiting on an insurance claim.  I also got a Hitachi V-1050F because it goes down to 500 microvolts and it arrived ok.

From what I have learned on this forum is noise is ok for digital work but not for audio.   Part of my learning will be what is acceptable and how to test the supplies.   Coming up with specs is part of this learning process.

I haven't been able to read the IC name but it's something like the LM2596. Look at that datasheet and you know more even if it's one of the Chinese copies.

I could not read it either but until I get them I am gong on the assumption that it is LM2596 and will start studying the datasheet.  I am trying to put an open source project together so I will need schematics and lots of details.

Edit: the 95% efficiency is the best you will ever see at one specific input/output voltage and current. Don't expect to have that where you want to operate.

I am not sure what you mean.  So I will take this as a learning goal - ie different % at different operating points.   Since my goal is to maximize the use of the laptop supply, I do have to pay attention to wasted heat.   I am a little concerned on the large (see above) module because it looks like a linear on a heat sink.  I hope it is for show only and most of the heat will come from the LM2596 (unavoidable heat).

I also have a space limit so I have to juggle all these modules to get the correct current (laptop supply limit) and space.  Now I am thinking of using the small ones for the set voltages and one big one for a variable source.  I think I will start with balancing the currents according to the specs and de-rate based on testing.

So it looks like I will start with 4.7 amps and divide that up into voltages - I am going to start a new topic - what voltages?

thanks

[ez24]

Here is one with heat sinks:


 [url=http://www.ebay.com/itm/DC-DC-constant-voltage-current-buck-converter-LED-Driver-/111299051606?pt=LH_DefaultDomain_0&hash=item19e9f0c056]http://www.ebay.com/itm/DC-DC-constant-voltage-current-buck-converter-LED-Driver-/111299051606?pt=LH_DefaultDomain_0&hash=item19e9f0c056[/url]

I like the LEDs and here is how I think this works:   

1.  turn both pots ccw
2. apply power to input
3. turn voltage pot to desired voltage (output connected to multimeter)
3. turn off
4. short outputs via a multimeter on amp setting
5. turn on
6. turn current pot to desired current
7 turn off - it is set

Now this is what I think will happen - turn on and the CV LED will come on.   Hook up a load that draws less than max amps and CV LED remains on.   Hook up a load that draws more than max amps or short out and then the CV LED  goes out and the CC LED comes on.

Am I right?

It seems like this would be good in a classroom environment.  The instructor could tell if there is a short in the output circuit by just looking at the leds  ??

[Rick Law]

thank you kjs
...

The real limit is the maximum power and that's 50W (VERY optimistic) w/o heatsink. You can put a tiny heatsink on top but the better way to cool it is a heatsink on the non-component side of the PCB (and a LARGE one). The IC chip is mounted on the leadframe to get the heat to the PCB and there are via holes which transfer it to the backside. At 50W that thing gets pretty hot.

I spotted a module with a tiny heat sink on the LM2596 and a slightly larger one on the back.  I never saw a heat sink on a pcb so I thought it was a joke and now I cannot find it.

this is a switchmode power supply and the switcher runs at 180kHz. This also means that the output is pretty noisy and not what you would expect from a real lab power supply.
...

...
thanks

Are you sure you have an LM2596? According to TI datasheet, the TI's LM2596 is 3Amp max.  Since it is suppose to go 5Amp, I would venture to guess this one is an XL4015.

KJS is right, you will have a hard time getting it to handle 50W without some kind of a heat sink.

I think KJS is right, no chance this board can handle 50W of power without some heat sink on it.
EDIT: said that already..

thank you kjs
...

Edit: the 95% efficiency is the best you will ever see at one specific input/output voltage and current. Don't expect to have that where you want to operate.

I am not sure what you mean.  So I will take this as a learning goal - ie different % at different operating points.   Since my goal is to maximize the use of the laptop supply, I do have to pay attention to wasted heat.   I am a little concerned on the large (see above) module because it looks like a linear on a heat sink.  I hope it is for show only and most of the heat will come from the LM2596 (unavoidable heat).

I also have a space limit so I have to juggle all these modules to get the correct current (laptop supply limit) and space.  Now I am thinking of using the small ones for the set voltages and one big one for a variable source.  I think I will start with balancing the currents according to the specs and de-rate based on testing.

So it looks like I will start with 4.7 amps and divide that up into voltages - I am going to start a new topic - what voltages?

thanks

Efficiency depends on input voltage, input current, output voltage and output current.  It is not a fix number.  95% is probably the best case.

What do you mean by "...start with 4.7 amps and divide that up into voltages - I am going to start a new topic - what voltages?"

With a buck convertor (XL5015 (EDIT, typo) XL4015 or LM2596), you have one output voltage that you select.  What current your load draws is what the buck will give it as long as it is within upper limit.  So, you select the output voltage, your load determines the current at that voltage.

You set the output to a specific voltage - one voltage only.  So, you can't "divide that (4.7amp) up into voltage"

Rick

EDITed a typo and what I thought was cut sentence but re-pasted back by mistake.

[ez24]

Thanks Rick

I saw one module that had an "X" on it (that is all I could see), so until I get them I will look at LM2596 and XL4015.

Sorry I did not make it clear I will use one module for each voltage, so if I have 3.3v, 5v, 12v and one adjustable voltage, I will use
4 modules.  Bummer I should have made this clear, I apologize for this.

The total current for all 4 modules has to add up to 4.7 amps.  I just to not know how to divide up the amps.  Just a guess here is what I could do (to make it a little simpler lets add up to 4.5 amps)

3.3v   500ma  (maybe too high)
5v    2 amps
12v   1 amp
Adj    1 amp

I want to set this up for young people and I think starting with 5 volts circuits is probably the way to go.  So I would use this module for the 5 volts (because of the CC and CV leds)


[url=http://www.ebay.com/itm/DC-DC-constant-voltage-current-buck-converter-LED-Driver-/111299051606?pt=LH_DefaultDomain_0&hash=item19e9f0c056]http://www.ebay.com/itm/DC-DC-constant-voltage-current-buck-converter-LED-Driver-/111299051606?pt=LH_DefaultDomain_0&hash=item19e9f0c056[/url]

and these modules for the other voltages (3 modules) because they are cheaper and smaller


[url=http://www.ebay.com/itm/201320736319?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT]http://www.ebay.com/itm/201320736319?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT[/url]

I have not calculated safety and power settings so I probably will have to adjust the currents or modules based on testing

thanks again Rick

[mariush]

Going a bit back to the basics, these kinds of regulators convert an input voltage to an output voltage with an efficiency that can be up to 95%. 

This efficiency is usually the value listed in the datasheet for the switching regulator chip used, in best case scenarios, with parts hand picked for a particular scenario, for example with a circuit designed for 12v input, 5v output, 2A maximum.

With boards like these that allow you to adjust the voltage, there are some compromises made in the components used around the chip, like the inductor and the diode, which means you won't really get that much high efficiency, estimate to have in reality an efficiency between 75 and 85%.  Not only this, but don't forget these are boards made to be as cheap as possible, so they won't use the perfect inductor choices, the best diodes etc, they'll use whatever's cheaper in quantity that day at the local market.
 
But what does this mean exactly?   Well, for example, it means that if you configure this circuit to output 10v and some device you connect to it uses 1A at this voltage, it means the circuit board with the switching regulator produced 10v x 1A = 10 watts. However, assuming the board did this with an efficiency of 80%, it means the board pulled about 10w / 0.8 =  12.5 watts from the input power connector.  Those 2.5 watts are lost as heat in the switching regulator, in the inductor, in the diode and to a much smaller degree in resistors and whatever else is in the circuit.

Now, if your input power is that laptop adapter, with the example above  it means that this circuit board alone used  12.5 watts / 19v = 0.65 A  ...

So now you know, for example as the previous post says, if you want a board to output 3.3v up to 500mA, it may do this with about 80% efficiency so that means the board may consume up to 3.3v x 0.5A / 0.8  = 2.0625 watts. With the laptop giving you 19v, it means the board will use at best  2.0625 watts / 19v = 0.108 A  from the laptop.
You can do the same calculations for the other boards and see if the total amount of current your laptop adapter can provide is enough.

It's important to remember that the efficiency of the circuit will be different at various output voltages and at various output currents. Keep this in mind when deciding to use several of such boards and power them from just one laptop adapter. Also consider this when it comes to keeping them cool - just because they say they can output lot of current, doesn't mean it's wise to use them at such high currents for long periods of times as they may overheat.

Also, while the seller on eBay advertises that the board can do a certain amount of current, you have to pay attention at the fineprint... i mean check the datasheet and look for notes about what's actually possible. 
It may be the case that - for example - the chip can do 2A in any condition but can only do 3 or 4 A provided that the voltage difference between input voltage and output voltage is less than 5v or something like that (so you wouldn't be able to output 3.3v at 4A with a 19v laptop adapter as input power)

[Didix]

What is the final goal of your project?

Why don't you just take a PS from a computer?

-Didix

[Rick Law]

...
The total current for all 4 modules has to add up to 4.7 amps.  I just to not know how to divide up the amps.  Just a guess here is what I could do (to make it a little simpler lets add up to 4.5 amps)

3.3v   500ma  (maybe too high)
5v    2 amps
12v   1 amp
Adj    1 amp

I want to set this up for young people and I think starting with 5 volts circuits is probably the way to go.  ...

Calculate WATTS, not AMPs.

When you convert 5V 1A from 10V, assume perfect efficiency, you would be drawing 10V 0.5A to output 5V 1A.  Of couse, nothing will have perfect efficiency.  Recall what Mariush said:

Going a bit back to the basics, these kinds of regulators convert an input voltage to an output voltage with an efficiency that can be up to 95%. 

This efficiency is usually the value listed in the datasheet for the switching regulator chip used, in best case scenarios, with parts hand picked for a particular scenario, for example with a circuit designed for 12v input, 5v output, 2A maximum.

With boards like these that allow you to adjust the voltage, there are some compromises made in the components used around the chip, like the inductor and the diode, which means you won't really get that much high efficiency, estimate to have in reality an efficiency between 75 and 85%.
.... ...

So, add up all the WATTS used and make WATT required at 1.5x used to 2x used - because your Laptop power brick may be listing "peak" so make sure you have head room.  Sustained max amp output may cook it.

---

If you are going to get a few, I would suggest get one cheap one first and get familiar with it.  That may help you decide which one you want 4.  Prudent to do before you get four that you may not like.

When you are testing it, use a low amp (300mA-500mA) 12V power brick so it doesn't put out that much power -limits the sparks and cooking of your board until you are familiar with how to adjust the board.

Typically, you have 2 or 3 pots, one for volt out, one for current limit, and the third one (if available) is saturation indicator (at what % of current limit should indicator turn on/off).  The current limit is not very accurate (10% range), and the saturation indicator is worst.

Typically, you will be "shorting it" with just an Amp Meter as load to adjust current limit and saturation%.  You may want to get some 0.1ohm current sense resistor - a bit safer to use than to risk your DMM.

I am a relative newbie as well.  I have been using a few of these boards as my main PSU.  They are fun to play with.

[ez24]

Update and without quotes

All the quotes are getting hard to read.

I got my first 90w adapter today and it is crap.  Much smaller and lighter than my Dells.  So it is out but someday I want to test it to the point I can make it smoke. 

Also I realize that if sell a brick it will have to be UL listed and this increases the cost 3-4x at least.  I will work on the bricks as time goes along.  I will allow the customer to buy their own brick if they want.

To answer someone's question - I do not want any AC main wires on the desktop.  I went to the library and looked at the children's laptops and they have adapters on the floor and I assume they are UL listed.  I want this safe for children, ie a children's bench power supply  again no AC wires on the desk

I will take the advice on using a bench supply to test the modules using the current limit from my bench supply.  It just came in as I was writing this, it is a Power Designs TP325, really cool looking (still looking for a 5005).  The volts check out but I do not have any leads to check the current.  Bummer   I will order some tonight.

I have a Wavetek 145, Hitachi V-1050F and the TP325  and I forgot to order any probes or leads   

I will take the advice on using 0.1 ohm resistors and ordered some 100 watters.   I was afraid to hook up my multimeter ( and it can only read 200ma).  My meter cannot read less than 1 ohm so I ordered a Uni-T UT139C.

I will not get any modules that use 3 pots, only 2 with V and I

Right now I am going to guess I will get from 50 to 70% of my 4.7 amps  and that will be a power supply with a total of 2.4 to  3.3 amps  (I do not know the watt setup yet)

I was going to buy some cheap Harbor Freight meters because I have blown up meters measuring current.  So the 0.1 ohm is a great idea - many thanks

The % has me thinking and the total currents and watts are going to be low.   Theoretically   IF  I have  a total of 2.5 amps what currents would you want (based on unknown uses and no losses) for

3.3 v
5v
12v

I took out the adj because of low watts and loses

I thinking of that a 4.7 amp ps maybe could do this :   3.3  = .5   5.0 = 1.5  12v = 1.0     
Since all modules are adjustable the user could lower one and raise another.  So I will need some sort of master warning LED.



thanks everyone

[Rick Law]

Update and without quotes

All the quotes are getting hard to read.

I got my first 90w adapter today and it is crap.  Much smaller and lighter than my Dells.  So it is out but someday I want to test it to the point I can make it smoke. 

Also I realize that if sell a brick it will have to be UL listed and this increases the cost 3-4x at least.  I will work on the bricks as time goes along.  I will allow the customer to buy their own brick if they want.

To answer someone's question - I do not want any AC main wires on the desktop.  I went to the library and looked at the children's laptops and they have adapters on the floor and I assume they are UL listed.  I want this safe for children, ie a children's bench power supply  again no AC wires on the desk

I will take the advice on using a bench supply to test the modules using the current limit from my bench supply.  It just came in as I was writing this, it is a Power Designs TP325, really cool looking (still looking for a 5005).  The volts check out but I do not have any leads to check the current.  Bummer   I will order some tonight.

I have a Wavetek 145, Hitachi V-1050F and the TP325  and I forgot to order any probes or leads   

I will take the advice on using 0.1 ohm resistors and ordered some 100 watters.   I was afraid to hook up my multimeter ( and it can only read 200ma).  My meter cannot read less than 1 ohm so I ordered a Uni-T UT139C.

I will not get any modules that use 3 pots, only 2 with V and I

Right now I am going to guess I will get from 50 to 70% of my 4.7 amps  and that will be a power supply with a total of 2.4 to  3.3 amps  (I do not know the watt setup yet)

I was going to buy some cheap Harbor Freight meters because I have blown up meters measuring current.  So the 0.1 ohm is a great idea - many thanks

The % has me thinking and the total currents and watts are going to be low.   Theoretically   IF  I have  a total of 2.5 amps what currents would you want (based on unknown uses and no losses) for

3.3 v
5v
12v

I took out the adj because of low watts and loses

I thinking of that a 4.7 amp ps maybe could do this :   3.3  = .5   5.0 = 1.5  12v = 1.0     
Since all modules are adjustable the user could lower one and raise another.  So I will need some sort of master warning LED.



thanks everyone

re: "I will take the advice on using 0.1 ohm resistors and ordered some 100 watters.  "

A 3W (5.57Amp) is almost an overkill.  You don't need a 100 watt 0.1ohm.  The 100 Watt is 31.6Amp!   Enough current to melt much of your other wires. 

re: "I will not get any modules that use 3 pots, only 2 with V and I"

The third pot (some calls it saturation, some calls it charge-complete) is not a waste if it is cheap and it typically is.  At times, I set it to lowest so it is an indicator if the DUT is beginning to draw current.  At times, I set it to 75% of current limit to show - hey, it is nearing max expected, I should double check.

re "... total of 2.5 amps what currents would you want (based on unknown uses and no losses)  ..."

Forget amps, use watts.  If your power source is 24V 2.5amps verses 16V 2.5amps, you are talking just 2/3 the available output at 100% efficiency.  Typically, higher the input voltage, better the efficiency.

re: "...I was going to buy some cheap Harbor Freight meters because I have blown up meters measuring current.  So the 0.1 ohm is a great idea - many thanks..."

You are welcome.  (Before I switch to a buck with digital display)  I use a pair of DT830B.  They are only $1 (plus $4 shipping).  One connected to a 0.1ohm as the current meter, and the other one connected as a volt meter.    Both the input to the buck and output from the buck should have its own 0.1ohm sense.  That will enable quick measurement of both and allows you to learn quicker.   I use gator-clips instead of probes.  With that as a set up, you can switch the pair from measuring input (from power source) to measuring buck's output to the load.  That will give you a fast understanding of input Watt verses output Watt (difference is the lost in efficiency), and V and I.

...Final thought...

Reading your OP and all of your replies, I think you would benefit a good deal working with one of them first.  Get a good understanding of how these kinds of boards works first before getting a pile of them.  Get to understand when your buck is connected to a load, your buck appears as a constant power device to the buck's power source.

You need to know the relationship between V, I, and Watt like the back of your hands.

Once you know it has enough power with the power source and individual module, how you limit the output power with multiple modules is really last thing to be concern about.  It would just be a trim pot adjustment - totally configurable.  So no need to be preoccupied with that just yet.

Make sure each module (for each channel) can carry the power intended for that voltage.  The biggest one you got is 12V 1A.  It is not so much power.  One with those bucks with mini heat sink should be fine with that.

When it comes to distributing power between modules, think Watts, not Amps.  You are dealing with constant power device as I said before.

Rick

[ez24]

Hi

I am starting to get my head around this.  I have several topics going on (on different subjects) and this is what I have come up with:

My project is a bench type power supply to be used in high schools in 3rd world countries.   I absolutely do not want any AC power lines at bench top level, thus using a laptop power supply that can remain on the floor.  No ATX ps, no commercial bench supplies.

Any laptop power supplies I furnish will have to be UL listed and I came up with this one:


[url=http://www.ebay.com/itm/Original-Toshiba-pa3516u-1aca-Global-90W-AC-Adapter-OEM-/280683893374?pt=LH_DefaultDomain_0&hash=item415a10527e]http://www.ebay.com/itm/Original-Toshiba-pa3516u-1aca-Global-90W-AC-Adapter-OEM-/280683893374?pt=LH_DefaultDomain_0&hash=item415a10527e[/url]

Because it is UL listed and has a standard 2.5x5.5 plug (very important)

It will be an option the school can buy their own laptop power supply (so I do not have to worry about UL)


I will use these modules:

http://www.ebay.com/itm/111299051606?_trksid=p2055119.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT


and


[url=http://www.ebay.com/itm/400714751657?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT]http://www.ebay.com/itm/400714751657?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT[/url]

I will test them to find their real specs using an infrared thermometer from Harbor Freight and datasheets.


I will use these outputs:  3.3, 5.0  and adjustable output

I will limit the current to 2.0 amps (each voltage) or less to protect the breadboards.   I also will need additional power to drive some other equipment (unknown right now)

I will learn how to use watts and power conversion calculations

I need to learn how to add additional protection to the 19v circuit  (maybe fuse) - cannot have enough protection

I need to learn how to build a negative tracking op amp ps  (I have no clue on how to do this)

I will use 0.1 ohm resistors to test the power supply currents (thanks again) by hooking them in series with the power supplies and measuring the voltages across the 0.1 ohm resistors.

Once I get it all figured out, I will try and raise funds through KS to come up with 10 units and try and find volunteers to assemble them and find a school somewhere to try and use them.  Then if everything works out the school can assemble the next set and pass those on to another school. (too idealistic ??) and so on.  This is a non-profit adventure, I am retired and this is fun to do.

-  will the ebay seller "f-t-2000"  be able to one source all my Chinese items ?  If this gets off the ground, it would be good to have one source.

thanks everyone, I got a good start
Roger in San Diego

[mariush]

No offense, but your level of knowledge is too low to think about Kickstarter and building things using funds from there.

Please try to learn about Ohm's law, relation between voltage, current, resistance, get some basic knowledge about how buck converters work ... for the moment stay away from opamps, negative voltages and other things, you're trying to do too much with too little knowledge.

My suggestion then would be to write down the requirements of your project and set them in stone. Don't write down too much. Once you decide on something, stick with it.

Another advice that I could give you would be to stay away from mains voltage and fuses and protections and all that by delegating that to existing, quality, certified power supplies.
Distributors of electronic components also sell AC-DC adapters of various voltages and maximum currents that actually respect and guarantee what they say, unlike adapters you find on eBay or at various dubious sellers.
For example, here's what AC-DC adapters they sell :

* US , Digikey: http://www.digikey.com/product-search/en/power-supplies-external-internal-off-board/ac-dc-desktop-wall-adapters/589898
* Farnell (Europe, Newark.com in US) : http://uk.farnell.com/ac-dc-converters-external-plug-in-adaptors-single-output

These adapters would have their own internal protections, fuses, you won't have to certify your product as it will only work with low voltages, 12v or 15v or 19v or 24v, whatever you decide to use in the end.

Another suggestion I would have is to make your device simply have an input connector at the back side where you plug such an ac-dc adapter and then in the front you can put a slide switch to flip between 3.3v out , 5v out , 12v out (if you want), adjustable voltage out (if you want)  and a potentiometer knob to allow kids to adjust the voltage.

Optionally, you could also add on front of the box voltage and current indicators, you can find such meters cheaply on eBay , here's an example for voltage or one for current and voltage. 

Note that such displays may need an insulated 5v power supply or something like that to run.

Such power supplies you can also find in a separate category at online distributors, for example at Digikey they're located here and Farnell has them here.
 
From the same online stores, you can also buy project boxes, made of plastic or aluminum, so you'd just have to do some cutouts for the output connectors and displays and selector switches and figure out a way to attach all the boards inside the box. With a bit of patience, the end result could look very professional. 

[ez24]

No offense, but your level of knowledge is too low to think about Kickstarter and building things using funds from there.

...Please try to learn about Ohm's law, relation between voltage, current, resistance, get some basic knowledge about how buck converters work ...

I agree - KS is a longs ways off

Distributors of electronic components also sell AC-DC adapters of various voltages and maximum currents that actually respect and guarantee what they say, unlike adapters you find on eBay or at various dubious sellers.
For example, here's what AC-DC adapters they sell :

* US , Digikey: http://www.digikey.com/product-search/en/power-supplies-external-internal-off-board/ac-dc-desktop-wall-adapters/589898

Did not think about Digikey.  I like their filters and by entering the same specs as a Toshiba power supply (at $25)  I came up with one for $35.  I will keep this in mind.  What I like about this are the spec sheets.

Another suggestion I would have is to make your device simply have an input connector at the back side where you plug such an ac-dc adapter and then in the front you can put a slide switch to flip between 3.3v out , 5v out , 12v out (if you want), adjustable voltage out (if you want)  and a potentiometer knob to allow kids to adjust the voltage.

Yes -  that is why I am using 2.5x5.5 mm plugs.  All kinds of female adapters are easy to find.   I will be using several modules that will be set by the instructor  ie V and C.   I will leave it to the instructor if he wants to give the kids little screwdrivers to turn the pots.  I figure they could determine the level of trust.

Optionally, you could also add on front of the box voltage and current indicators, you can find such meters cheaply on eBay , here's an example for voltage or one for current and voltage.

I will be doing this and have bought every kind I can find to evaluate later once I get the power modules nailed down.

From the same online stores, you can also buy project boxes, made of plastic or aluminum, so you'd just have to do some cutouts for the output connectors and displays and selector switches and figure out a way to attach all the boards inside the box. With a bit of patience, the end result could look very professional. 

This is a big deal and I am keeping my eyes and mind open.  But I want students to be able to put together these things for other students (at the end of the course), so cut outs are probably out of the question (sharp tools).

It will be a challenge mounting the modules.  I have been experimenting with different methods, using screws, bolts, brass standoffs, plastic standoffs, even super strong velcro, plastic ties, and glue.  I am going at this trying to mount them to a plastic surface (plastic boxes) without drilling anything.  I have not found a good way yet.  The most promising method is using strong velcro by using different sizes, ie larger on the plastic and smaller on the board.  That way when it is pulled loose the one on the plastic remains.  I have thought about gluing a dense plastic or light wood to the plastic case and just screwing into it.  It is tuff trying to do something without drilling.  I think mounting is going to be the most difficult part.   I want students to be able to put these together without power tools.  One method I am going to look into is using hot glue.  A hot glue gun could be included with the units.  I probably should start a new topic on fastening to plastic.

Another crazy idea is using a breadboard for the support.  I have had somewhat good luck mounting breadboards to plastic (because of large surface) using the strong velcro.  I am looking for breadboard to standoff connectors (probably none exist).  Also looking at using headers that fit a breadboard and trying to figure how to use them to mount a pcb.  Probably a good topic.  whew....

A forgot about the breadboard power modules that I got ..