You don’t have the power? Well, there’s no need to ask Scotty – the SparkFun Benchtop Power Board Kit has your back. This board will let you take advantage of your old (or new) power supply to create a benchtop power supply with enough juice to run almost any of your embedded electronics projects.
The benchtop power board kit was created to provide quick access to the typical voltages needed when developing physical computing projects (embedded systems). After assembling the kit you'll have access to four different voltages (3.3V, 5V, 12V and -12V) each with their own replaceable 5A fuse. Each power rail has a corresponding ground connection; all of the power rails are brought out to a binding post. The benchtop power board should be powered by a standard computer power supply with an ATX connector.
This kit is simple to put together and shouldn't take more than 30-45 minutes for a beginner.
This skill defines how difficult the soldering is on a particular product. It might be a couple simple solder joints, or require special reflow tools.
Skill Level: Rookie - The number of pins increases, and you will have to determine polarity of components and some of the components might be a bit trickier or close together. You might need solder wick or flux.
See all skill levels
Whether it's for assembling a kit, hacking an enclosure, or creating your own parts; the DIY skill is all about knowing how to use tools and the techniques associated with them.
Skill Level: Noob - Basic assembly is required. You may need to provide your own basic tools like a screwdriver, hammer or scissors. Power tools or custom parts are not required. Instructions will be included and easy to follow. Sewing may be required, but only with included patterns.
See all skill levels
If it requires power, you need to know how much, what all the pins do, and how to hook it up. You may need to reference datasheets, schematics, and know the ins and outs of electronics.
Skill Level: Competent - You will be required to reference a datasheet or schematic to know how to use a component. Your knowledge of a datasheet will only require basic features like power requirements, pinouts, or communications type. Also, you may need a power supply that?s greater than 12V or more than 1A worth of current.
See all skill levels
We welcome your comments and suggestions below. However, if you are looking for solutions to technical questions please see our Technical Assistance page.
No reviews yet.
This board starts to smell really really bad at around 15 amps (If you jump the fuse).
These don't look like they're spaced at standard banana jack spacing. Isn't that important anymore?
Agreed! How am I supposed to connect my .75" adapters?!?
its low current (5 amps) which shouldn't be enough to spark across the terminals at those voltages
Sparking ability has nothing to do with current.
I think the point is that most double banana plugs have a spacing of 0.75". The banana jack spacing on this board appears to be about 0.5".
This is a great idea! It is too bad these don't have a power switch. If the ATX connector was on the back and these had a power switch, these could be mounted, using standoffs, on the front of an enclosure. I like not having to modify the power supply.
Exactly, turning this into a fully enclosed "unit" would be a wonderful idea. How's this for a tag line:
"BYO ATX PSU"
Hmm...too many acronyms.
LOL UCWIDT?
Nice concept, Needs work. A lot of my points have already been mentioned by @ctdahle but still have not been fixed/addressed. 1) Standoffs are not long enough. I have attached mine to the PSU frame, and had to trim off the ends of the banana terminals in order to keep it from touching the 'floor'. 2) Would be nice if you included 8 screws for the standoffs. Not fun rummaging through my screw bin to find 4 more screws so I can attach the standoffs to something. 3) Banana terminals are cheep. It is very easy to strip the plastic threading on them when trying to tighten them I ended up using blue lock-tight and super glue to keep them from moving. Which brings up point 4... 4) Since the board is machined, why is the holes for the banana terminals round? The Banana terminals are double-d (Oval shape with 2 flat sides) for the intention of them to stay put and not spin freely when tightening? 5) Termials are too close together. I have small fingers and I had a hard time using, including tightening the nuts on the banana terminals. Why are they not spaced at the normal banana terminal spacing? 6) Wires for terminals are also to close to the holes. I had to add hot glue and heat shield to prevent nuts from touching the nearby wire, or terminal next to it. 7) Include Eye-terminal crimps with the kit. I shouldn't have to go digging though my supply to come up with additional parts for this kit. Wraping bare wire around the terminal that connects to the board will not work. You NEED eye-terminal crimps. 8) Added perk. Since you are allowing us to connect a ATX connector. Give us the full potential! I would recommend making a 'deluxe' version of this product to include -5vDC, On/off switch (PW_ON -> COM), and Power_ok Light.
I like the product and it's concept and am ok with what I currently have. But for something that has been here for more than 3 years. This device needs an overhaul bad.
I had to buy new Banana Jack posts from Radio Shack to make this work right. The posts that are included are very cheap quality. I ended up getting 274-0661 Insulated Binding Posts. These are significantly shorter and have an all metal mounting post. The new posts themselves are thin so I filled the original mounting holes in the board with 2 part epoxy (put tape on bottom to act as molds). Then I drilled holes to fit the posts exactly. I put those in the board and used the tabs to solder to the holes. Another issue with these boards it is exceedingly difficult to solder the grounds. I ended up cranking my iron to 800 def F and using lots of flux to get those to solder properly.
So between running to the Shack and fixing the board to make it usable I would probably just rig up my own solution in the future. Like eqrunner said, these boards need a re-think.
This is one of the most useful kit's I've bought. My only criticism (as others have mentioned) is that the standoffs are about 1/4" too short to be safely usable.
I bought this in December, assembled it in January and used it on and off for the past few months, but I'm not real happy with it. In fact, it's the only SparkFun purchase that I've ever been disappointed with.
I had previously hacked an ATX power supply to construct my own bench top supply and have been using that hacked supply for a while. Needing a second DC power supply, I thought this kit would be a quick and easy way to get one with out the hassle of hacking apart and modifying the case and wiring of another supply. Regrettably assembling this kit and getting it running did end up requiring more time and effort than I took hacking the first one from scratch.
First issue: None of the 3 "ATX" power supplies I harvested from old computers had connectors that matched this kit...it has a 24 pin connector and all of my supplies had 20 pin harnesses. At the time I bought the kit, I was too much of a tyro to know that "ATX" was not a generic power supply and that different "ATX" supplies would have different output harnesses and pinouts.
I was able to find a pinout map my 20 pin supplies and for the 24 pin supplies that this kit was made for, and eventually work out how to connect this, but doing so required ripping apart the harness on my supply and then individually fitting each of the pins into this board's connector.
Second issue: The provided stand-offs are not long enough to support the board. As a consequence, the terminals on the back side of the binding posts interfere with the base plate to which I had planned to mount the board. Fortunately I had some longer stand-offs on hand, but I buy kits so I don't have to rummage around for parts and pieces.
Third issue: My supplies all do need a load on the 5v rail in order to stay latched on. It would be better if this board was made with a location to solder in an appropriate load resistor if needed. I stuck one on, but it looks ugly.
Fourth issue: Given the VERY high quality of other goods I've bought from SparkFun, I was shocked at how poorly made were the supplied binding posts. Not only are they poorly molded, but the nuts on the back side were not accurately sized to fit the threads of the screw terminals. Thus it was very difficult to snug them up to the circuit board adequately so that they don't fall off when I connect wires. I ended up snugging everything up as tightly as I could and then drenching the fasteners with HotStuff (cyanacrylate "super glue") hoping that it would keep the biding posts from detaching from the board. Additionally the supplied binding posts are a very loose fit for the variety of banana jack leads I have on hand. An inadvertent bump or tug is sufficient to disconnect the leads, leading to borderline OCD behavior as I found myself constantly checking to see if the power was REALLY connected to various projects.
In the end, I finished the kit out of pure stubbornness, satisfied my self that it worked, and then shelved it. Still needing that second supply, I went ahead and hacked up the ATX this past weekend and built up a little LM317 variable regulator to enclose within it's case to boot, a project that took less time than assembly of this kit.
In short, I'd say this kit is worth it if you happen to have exactly the same computer PS that this kit is designed for, but I suspect it won't satisfy most hobbyists or experimenters for long. Find one of the many online tutorials for hacking your supply and do that instead.
20-pin ATX plugs are a strict subset of the 24-pin plug. The 20-pin plug will fit in one end but not the other. Done. Works. I've powered multiple computers this way.
This is a link to an article discussing converting an ATX PSU to a benchtop supply and the suggested load resistors.
http://www.techbites.com/201004242624/myblog/blog/z0028-further-to-benjamin-jordons-atx-psu-to-bench-top-supply-conversion-project.html
Recommended load resistors - from article
5.6 ohm, 5W - 3.3v
15 ohm, 5W - 5v
12 ohm, 25W - 12v
The min load on the 12v supply is 1A. Some PSU's won't meet spec on any rail if the 12v rail isn't loaded. This article disusses post regulating the 12v rail rather than a static load (read space heater), an LM317 might work here.
Hmm. Could a spool of red wire and a spool of black wire be added to the "Related Products" list for this? :)
I think i'll stick with modding my power supplies the old fashioned way. This is an awesome product, but even a 200 watt PSU runs well past 5 amps, and its nice to have the power there is you need it. Still, this is a great idea for hobbyist who don't want to ruin the pc usability of a psu.
Dont you need a load resistor on a line? Your PSU will go poof without one on a 5v or 12v line. If it doesn't go poof, it will refuse to supply power.
That is true, but different PSU's require different minimum loads. Mine for example requires 0.3 amps or something on the +12V line but others require stuff on the +5V line. Also, it'd probably be difficult to get the heatsinking right on a load resistor mounted on this PCB.
The ATX specification gives the minimum current on each line. The kit ought to come with loads that meet those requirements.
Terrible design. The holes for the binding posts are round, and not d-shaped, meaning that as you tighten the posts, they will start to turn and loosen. I'm rather disappointed at this obvious design miss... and a big miss it is. Just for this serious design flaw, I'm out.
This Board works Great.. I Got an Small Power Supply from an old PC and I put it to a Radio Shack Enclosure box and I built a nice Power Supply... Really Recommended...
i ordered one of these a long time ago, and just recently started looking at it with an eye towards assembly. although fuses sre funtional, i know im gonna blow an ungodly amount of them., now seeing as the standard atx supply has short circuit protection. does anyone have a circuit i can use to replace the fuse( like some sort of fast acting crowbar to trigger the internal shutdown of the psu
Why not use a resettable fuse to (try to) protect this first, rather than intentionally freak the PSU out? :)
Oh man, where was this the other day? I found an old mini ATX power supply in the closet. I tossed it out thinking I had new use for it. I so could have made it into a benchtop power supply.
could be better. The fuse holders are keyed... but the keying is so frigging tiny that you're likely to miss it and then waste 30+ minutes unsoldering the holders. The holes for the wire you need to add to get power/ground to the connectors are tiny, tin those in advance and you'll never get the wire through (another 30 minutes lost)... and then at the end, you might notice that the standoffs that the board is supposed to sit on are exactly the same length that the connectors protrude from the bottom of the board... so you'd be wise to seal that up to avoid touching down on anything conductive... in short, a number of minor issues that wouldn't be so annoying if they weren't so trivially easy for sparkfun to fix! :(
Anyone have suggestions about a circuit breaker type addon to replace the 5A fuses? I'd much rather have to push to reset a circuit breaker (even if it costs a bit more) than having to replace fuses over a mis-wired component.
The description makes this sound like your one stop shop to use your old ATX PSU as a benchtop supply. Not so, your average PSU will require at least one dummy load for the PSU to stay on and for the voltages to be in spec. Unfortunately I bought based on the description and didn't take a look at the schematic. Now I'm ordering $4 worth of components to hack it into a working solution.
At the very least the description should give some indication that your PSU may not stay powered on or the voltages may be far out of spec without one or more dummy loads. Mine seems to only require a load on +12v. Others may require loads on 3.3 and 5v too. I'm buying parts to load all 3 so it'll work with any PSU.
Please change the design, include the $4 worth of dummy loads, or at least update the description. The comments section looks like 2 years worth of good number of people having issues with this product.
Is the 1k resistor and the LED considered the load on the +5V rail.
Depends on the PSU's requirements. A 'safe' number for the 5v minimum load is 300mA (much more than the LED). A PSU may require minimum loads on multiple lines though. Typical PSUs may need 1A on +12v, 0.3A on 5v, and 0.5A on 3.3v. If there are multiple +12v lines then each may require a 1A load.
What a great idea, what a bad board... I got the board and noticed, that it is barely accepting solder. Finally the connections, where the cables are soldered, did not accept soldser at all. The board is a total waste of time and at the end, I decided to put it where it belongs: wastebasket. Sorry, but this is not the quality I am used from other PCBs of sparkfun Ciao, Mathias
I think it's partly the finish on the board and partly just the heatsinking due to the sizes of the traces. The ground wires were a real pain and the location of the solder points for the binding post wires are inconveniently between the binding post and the fuse clips...
I'll just leave this here :)
http://www.thebestcasescenario.com/forum/showthread.php?t=25018
Some things I noticed, and aggregated improvements from other comments for a future revision:
1) Use a "D" punch to keep the terminals from rotating, and to keep the wire holes accessible. I had to fiddle with them to straighten all the posts to do this.
2) Consider increasing the standoff to keep posts from interfering or shorting to a conductive surface (if mounting on PSU) and so the entire post is available (if using the box). One way to do this for the box without modifying the PCB layout is to mount the connector and fuse clips on the "bottom" of the board and include longer standoff posts. This should bring the posts up enough to access the wire holes I think. These look like through-hole plated so I think this would be possible, but please correct me if I'm wrong.
3) Include enough fuses! I bought three kits that only had two in each.
4) Make a help page to aid users in determining if their old PSU is compatible with this kit (types of connectors, figuring out if it needs a load, etc.)
5) Space the +/- posts 0.75" apart for the standard "double" banana plugs
6) Minor: include some red and black wire in case someone doesn't have it on hand
Is pin 3 really a ground? If I follow my MM it lines up with the +12V which isn't making sense to me.
I just checked, and I believe the pinout is not backwards. You should be fine.
So a lot of these comments are about how this board is not necessarily compatible with all power supplies, whether it's the need of the load resistor or whatever.
However, can I assume that this will work properly with the power supply sold here at Sparkfun? (TOL-09539)
Really cool, but it really needs a power switch.
This is great, works very well. Missing the LED from my kit, but no big deal really.
One solution to an ATX PS that senses load before turning on, is simply to plug in an old CD drive or other computer hardware. I chose a CD drive because it's silent when not in use, unlike an HDD. Not as elegant as a power resistor, but it got the job done with parts on hand.
Eagle files please
agreed, eagle files would be nice
Eagle files posted.
Most hobbyists have an ATX supply sitting around doing nothing so this kit is a good value. There are plenty of YouTube videos on this subject, but the SFE kit is quicker, safer, and more professional.<br />
Not spacing the binding posts on the standard footprint is a problem that the manufacturer should correct.<br />
It is true that most ATX supplies need a load to produce the rated voltage, but I have never heard of any going "poof" without a load, they just go into standby.<br />
If I get time, I will draw up a dummy load circuit that adjusts for how much power is being used externally. That is: if you are using power, the dummy load will adjust so it is not making extra heat and wasting watts. Considering the whole point is for this thing to be an inexpensive alternative, you can't get too elaborate with the bells and whistles or you might as well buy a budget lab supply instead. The switch is a great addition, easy, and cheap too.
One other thought...
The back of the banana plugs bottom out when the unit is placed flat. If this were set on a conductive surface while power is applied, bad things would likely happen. Having some longer standoffs would be a bit safer.
Just bought one and put it together. IMHO the only thing it's missing is a switch. I ended up modding a switch into the pcb by breaking out pin 16 ( PS_ON - green wire ) and switching it to common ground.
Not all Power supplies ( especially older ones ) have a switch on the back. Plus I wanted to set this up on my workbench and have the PSU hidden out of the way. Having to unplug either the PSU or the connector seemed silly.
The only other thing worth noting is that since the grounds are all common, three of them are redundant, so I left them off. I am using single banana plugs, so I just move the red one around as needed.
Great idea. These power supplies are so cheap for what they do.
I'd like to see another generation that was built to handle more power and included the load resistors. The ATX supply that SF sells is rated for 15A at 12V. So how about a 10 amp fuse for that pole? Also the traces are pretty small for moving that kind of power.
My math comes out to 25, 33 and 24 ohms for 5, 3.3 and 12 volts respectively, which is even better.
If only there was some magical way to search for info like this, perhaps using a series of tubes.
12V / 0.600 A = 20 ohms and 5V / 0.200 A = 25 ohms, but you need to account for the resistor tolerance. Rounding those down to the next standard size give 17 ohms and 22 ohms.
You could go with a 30 ohm 1W resistor on the 3.3V, maybe even a 1/2W, but that would be pushing it.
Current ATX spec at http://www.formfactors.org/developer%5Cspecs%5CPSU_DG_rev_1_1.pdf calls for minimum loads of 600 mA on +12v, 200 mA on +5 v, and 100 mA on +3.3 V, and that no damage result from underloading the outputs.
Based on that, 22 ohm 2W resistors should do for the +3.3 and +5 v rails, and 17 ohm 10W for the +12 rail. Other rails must permit 0 mA minimum load.
I have one of these with a 15ohm 5w on +5, 7.5ohm 5w on +3.3 and 15ohm 25w on +12v and it works, but the 12v resistor got hot as the bejeezus belt. I removed it and my ps (Enermax EG-365) still comes on, but the 3.3 supply is only 2.8v. I intend to use this mainly for high current 12 volt situations (charging li-polys) so it won't be a factor there.
Has anyone else actually bought this? I seem to be having problems with the pinout, as it looks backwards.
Yeah, I traced the connections it looks like the pinout is not backwards.
kinda expensive for what it is i think it should be maybye 10$ then i would purchase it.
order 100 of them and they will be about 10 bucks each haha
What about some caps to even the ripple?
I think it's assumed that your power supply on your project would have proper filtering.
I wish SFE had this a few years ago. I modified a 350 Watt Antec PSU to get the same feature. I know I spent more than $14.95 in parts too. This would be nice - no cutting or splicing or trying to stuff all of the wires into the case without shorting the rails - and it doesn't take part of a day to make it look nice!
Just check the supply you want to use with it. There are plenty of PSU's that have a very low minimum load. I skipped the load resistor on mine and it will still source very small circuits - and it hasn't grenaded even after long periods of use.
I've been using ATX power supplies for my projects for years without hearing that they are supposed to have minimum loads. I've never had any problems. It is very easy to find a powerful, used power supply for $10, though, so until they start dying on me I think this is a risk I'm willing to run.
Years ago it used to be a problem - usually the switcher wouldn't start if there wasn't a minimum load. But I haven't seen a PSU with this problem in at least 4 years.
I have personally used old supplies as well without an issue. I didn't realize a load was needed either, oops! It just depends on the supply.
I would further agree with the other comments. This might work with some computer power supplies, but I'm betting that a significant proportion of users will end up having problems if they don't add something for minimum load.
I'd also second the note on double banana plugs and go on to mention that they would be most useful if they were arranged to provide GND and Voltage with each double-banana connection. One other way to go about it would be to make an alternative version of the board that would wire to panel-mount banana jacks. That way if you put everything into your own choice of enclosure, you don't have to figure out how to deal with the extra large holes. It may be possible to do that already, but without a picture of the bottom, it's unclear how the jacks connect with the board.
Just my thoughts. Otherwise it's a great product idea.
Despite the lack of a built in minimum load (like what is used in most PSU testing circuits, which I think use several-Watt-dissipating resistors, right?) and the lack of a power switch on the unit itself, this is still a great idea and I hope it continues to be polished.
Most likely 5V, with a 1k resistor, 5-1.7V = 3.3V. 3.3V/1k = 3.3mA. Not much, but probably enough to keep it from poofing.
The LED would provide a little load for one of the outputs. I don't know which output it is connected to though.
The LED is connected to the 5V outputs, in series with a 1K resistor, according to the schematic.
hmmmmmm: Dont you need a load resistor on a line? Your PSU will go poof without one on a 5v or 12v line. If it doesn't go poof, it will refuse to supply power.
Agree. This is not a 'benchtop power supply' but an ATX power supply killer.
I'd expect at least a warning in the assembly guide.
That's what the fuses are for.
Wow, I need exactly this. Cool!
Awesome!