Powering large breadboard setup

[5321488]

I am trying to power 17 logic chips over 4 breadboards (see picture), mostly 74LSXX, but a few CMOS and 3 555s, with a stripped USB cable plugged into a 5v output phone charger. I can't get everything to turn on - whatever is close to the power will work but not other stuff. I measured 4.6-4.9v at the power strip where it is plugged in, and then as I measured each strip further and further away, it dropped to around 0.8v, obviously not enough to turn things on.

I'm not sure what the problem is,, particularly given that I've seen people power far more hardware also with a small plug pack. Is it that the connections in and between boards have too high a resistance and the voltage is dropping? Is it just that I can't supply enough current? Do I need to try a beefier supply, regulator and capacitors across power, separate supplies for each board?

Thanks !!   

[Audioguru]

Get rid of the breadboard with its intermittent contacts and the messy tangle of wires all over the place. Replace it with a soldered well-designed stripboard or pcb that has a compact layout.
Old TTL uses a fairly high current and ordinary old 555 ICs are the same, use all Cmos digital and Cmos 555 ICs.

[5321488]

Thanks for your help! That sounds like the way to go, although a shame that the TTL isn't viable.

[ez24]

Use more wires

[casinada]

1.- Probably you need a more powerful power supply.
2.- Use decoupling capacitors on each chip.
3.- Use thicker wires for power or try to power each breadboard directly from the source.

[Rick Law]

I am trying to power 17 logic chips over 4 breadboards (see picture), mostly 74LSXX, but a few CMOS and 3 555s, with a stripped USB cable plugged into a 5v output phone charger.
...
...
I'm not sure what the problem is,, particularly given that I've seen people power far more hardware also with a small plug pack. Is it that the connections in and between boards have too high a resistance and the voltage is dropping?
...
...

From one newbie to another...

Sounds like you need a DMM.  As you trace down the breadboard(s), you can see how the voltage drops.

Even a cheap DMM will help you track down where you may have a breadboard hole that may not be making the best of contacts.

[Ian.M]

I would suspect the jumper wires.  Cheap and crappy Chinese ones are often made with wire that has a very thin metal core and may not even be copper.  If their resistance is too high then yes, you will get trouble like that.

I have no problems with breadboard power distribution, but I don't use pre-made jumper wires.  Instead I use wire stripped from Ethernet or Telephone cable - 24 AWG SOLID COPPER with colour coded insulation.  (Don't use Ethernet patch leaad cable, its stranded and will FUBAR your breadboard.)  Each wire is individually cut to length for a neat layout, and when running jumpers between breadboards in a panel to hook up the power bus strips, I put them at both ends of each breadboard.  That keeps the resistance as low as possible.

Most signal wires are less critical - if you don't expect more than 10mA through it your jumpers should do, but please check their resistance with a DMM before use to avoid unpleasant surprises (e.g. badly crimped end pins that may barely be making contact).

Here's a VFD clock I built a few months back:

That's the Arduino, four chips, the VFD filament, and a 24V output DC-DC module powering the 12 transistors driving the VFD, all running off the breadboard bus strips distributing 5V from a hacked phone charger (Old phone plug cut off and replaced with switch on left with LED + red and black wires to pins for breadboarding).  The 5V rail is within 0.1V of the input voltage everywhere.

[james_s]

Some USB cables have very thin wires, originally USB maxed out at 500mA. Many USB adapters meant for charging phones and such have awful regulation, the output is noisy and it's all over the place so make sure you have a good regulated source capable of delivering sufficient current. As someone else mentioned, you will want to have a generous sprinkle of decoupling capacitors, a lot of TTL devices back in the day used a separate 0.1uF capacitor across the power rails right next to each IC.

[Ian.M]

Yes, I forgot to mention decoupling.  The switch 'pod' also has a 100uF cap inside it on the PSU side of the switch, and on the breadboard I've got 0.1uF directly accross the power pins of each logic chip, 47uF between the 5V rail and the 0V rail and 10uF on the 29V rail feeding the VFD drivers.

You can check if your USB supply is 'fit for purpose' by loading it a bit with a 100R resistor and connecting your multimeter with a 1uF cap in series with it across the supply, and reading the ripple on a low AC volts range.  See my advice from 'Note 2:' onwards in Reply #97 here.

[mariush]

Buy boards like these: http://www.ebay.com/itm/5-x-Prototyping-PCB-Circuit-Board-Stripboard-Veroboard-94x53mm-Copper-Free-Ship-/281765908639?hash=item419a8e909f:g:iJIAAOSwjVVVwYB4



Same stuff here: http://www.ebay.com/itm/5-pcs-PCB-Prototyping-Circuit-Board-Strip-Board-100x50-mm-Free-Shipping-/262007892123?hash=item3d00e3289b:g:WR8AAMXQqdhR68sf

And these may actually be more value for you and your particular design, they're more like prototyping boards in the layout of the traces in the back (and you get 10 boards for less money): http://www.ebay.com/itm/10Pcs-Lot-DIY-Prototype-Paper-PCB-Universal-Matrix-Circuit-Board-8-5x20cm-/281761355350?hash=item419a491656:g:-s4AAOSwkNZUlSHB


You can use a sharp blade to cut the strips where you want (when I'm lazy i sometimes just pick my a phillips driver or a drill with phillips screw tip and just enlarge a hole, breaking the narrow copper sides)
You can add a layer of solder to strips to make them carry more current. For even more current capability... I sometimes take solid core wire (like the leads of components) and route it through two holes on the same strip but with the wire on the copper side, and then solder the wire to the strip over the whole length of the wire.
Another advantage...  you can add ceramic decoupling capacitors (0603 or 0805) directly between two strips or where you solder the pin of a chip so you don't load the top side with components.
You can make connections between strips directly using solder on the bottom side and between more distant strips with wire from the top side (when i'm too lazy to go pick my spool of awg24 wire, i use pieces of resistor leads or i sometimes strip a bit of ethernet cable to get wire from it, but lately this isn't such a good idea because a lot of cheap ethernet cables are just copper clad aluminum, lousy wire)

[rstofer]

On some breadboards, the power and ground strip are intentionally interrupt half way down the board.  Check on both ends of each power/ground strip.

Right off, I don't know how to identify which boards have this kind of strip so I use my meter.

[5321488]

I am trying to power 17 logic chips over 4 breadboards (see picture), mostly 74LSXX, but a few CMOS and 3 555s, with a stripped USB cable plugged into a 5v output phone charger.
...
...
I'm not sure what the problem is,, particularly given that I've seen people power far more hardware also with a small plug pack. Is it that the connections in and between boards have too high a resistance and the voltage is dropping?
...
...

From one newbie to another...

Sounds like you need a DMM.  As you trace down the breadboard(s), you can see how the voltage drops.

Even a cheap DMM will help you track down where you may have a breadboard hole that may not be making the best of contacts.

Thanks! I have traced the strips with a DMM, there doesn't seem to be any missing contacts, just a large voltage drop each link from board to board. I haven't yet measured the current.

[5321488]

Thanks everyone for the suggestions!   

So: I'll get a more solid power cable/supply, stuff the jumpers, test the supply, directly hook up power to each board and decouple the chips. Hope that works.


(I have been using all 22AWG copper wire for power, and 22AWG, 18AWG and jumpers for more signal based stuff. I just trialled some cheap jumpers to hook up additional power, and sure enough they started to heat up and were high resistance.)

[eblc1388]

Use a magnet and test every jumper. Throw away any that sticks to the magnet and any jumper wire that feel "springy".

Recite "I should never judge a wire by its outside diameter" and say thanks to clever mechants from China. 

[Ian.M]

In addition to the advice above which eliminates steel wire, you also need to eliminate CCA (Copper clad Aluminium).  Although the conductivity of pure aluminium is only about 60% of that of copper, that's not the problem.  You don't want it in any flex cable or data cable because its tensile strength is much lower and it has nowhere near the fatigue resistance of commercial grade copper wire + it is very difficult to terminate reliably.   Measure your jumpers to find the highest resistance one (that passes the magnet test) then strip it and sacrifice it to the flame test!


N.B the flame needs to be yellow as show, not a fully oxygenated blue blowtorch style flame as that can (just) reach the melting point of copper.

[Zbig]

Here's a VFD clock I built a few months back:

Now, you can't really go wrong with something called "Bimboard", can you?

[Audioguru]

I have made many projects on stripboard. They all worked perfectly and the prototype looked good enough to be sold as the final product. Some were extremely large and complicated and used LS-TTL logic ICs but most used Cmos or opamps. I designed the circuit schematic then made a layout of the parts and a few short jumper wires on grid-paper. The copper strips were cut to as short a length as is possible with a drill bit.

[Rick Law]

...
...
From one newbie to another...

Sounds like you need a DMM.  As you trace down the breadboard(s), you can see how the voltage drops.

Even a cheap DMM will help you track down where you may have a breadboard hole that may not be making the best of contacts.

Thanks! I have traced the strips with a DMM, there doesn't seem to be any missing contacts, just a large voltage drop each link from board to board. I haven't yet measured the current.

Thanks everyone for the suggestions!   

So: I'll get a more solid power cable/supply, stuff the jumpers, test the supply, directly hook up power to each board and decouple the chips. Hope that works.


(I have been using all 22AWG copper wire for power, and 22AWG, 18AWG and jumpers for more signal based stuff. I just trialled some cheap jumpers to hook up additional power, and sure enough they started to heat up and were high resistance.)

re: just a large voltage drop each link from board to board

I think you just got the answer.

I personally use 22 to 24 AWG and strip them myself.  Except for power rails, I do on occasions use those pre-made jumpers mostly because I have a bunch laying around.  When I do use them, I am keenly aware they are not so great - I would occasionally double up a pair of jumpers to decrease the issue of bad contacts/high resistance.