DIY Electronics

[robk]

We need more do-it-yourself electronics projects here! This hobby should be more than just machining and direct drive, we need to test boost/buck regulators and innovative LED drive circuits. LEDs are unlike incandescent bulbs, they require current regulation for maximum battery life. Lets share information about new circuits, new designs, etc. Check out my down-converter post in this area, I can post the PCB image for others to use for etching. I've bought Badboys and Downboys, but it's cheaper and more fun to do your own.
Rob

 

[Doug Owen]

I fully agree, rub is making it happen.....

Doug Owen

 

[PEU]

This is a compilation of available step-up circuits I found online.

I'll look for these here so I can start something also, besides that I posted my 1st mod in the mod forum, here is the link.

My last post there is an explanation of the circuit used.

Pablo

 

[Ocelot]

Pablo,

You seem to be missing the LTC1308B step-up voltage regulator. I've used this device in some circuits I've done to drive a 5W Luxeon, and it works great.

This boost converter has a very wide input and output voltage range (1-10V in, up to 30V out).

The really nice feature is that it has an on-board LBI input which compares to an on-board 0.2V reference. This feedback path can be used to sense the current going through the load to convert it into a current regulator, and since it only needs 0.2V across the sense resistor, a lot of power isn't dissipated in the sense resistor (0.2W @ 1A).

I picked this converter based on the low input voltage capability and the on-board comparator. It seems to work very well in practice, too.

Scott

 

[Jack_Crow]

R,
Would love to, but my shop is 7k miles away.
There are no Radio Shacks here in Iraq.

Perhaps when I get home in December.

Hope all is well
Jack Crow

 

[PEU]

My reliable source for any of these circuits is also 7kmiles away... (digikey) /ubbthreads/images/graemlins/frown.gif

Pablo /ubbthreads/images/graemlins/smile.gif

 

[UK Owl]

How about this for a simple constant current regulator

For R2 use a large value resistor e.g. 1+ Meg
For your NPN transistor use ANY general purpose small signal transistor, but note that a lower dropout voltage for this circuit will probably be seen if you use a germanium.
The LED can be whatever load you require, from a single White LED or even a Luxeon.
R1 is your sense resistor and you will need to experiment to find the one you require for the current you want, this may also vary slightly depending on your choice of NPN transistor.
The FET will be chosen to carry the current you require, note: I have made this circuit several times driving one or two White LED's in parallel using a single ST72A FET (I got a bag of 50 on e-bay for £1 & £1 P&P).

This uses very few components, thus the circuit can be made very small. Versions I have made are very linear, and have a dropout voltage of arround 0.5 volts (using a silicon NPN), this is much less than that seen using an LM317 in constant current mode.

Hope this is of interest to someone.

 

[Neg2LED]

Look at my SIG!!!! it is full of a real cool step up circuit and the link to the site it is on! its called the Joule Theif, and im gonna be sellin them soon!!!!

 

[HarryN]

Thanks for posting your circuit.

As an example, if I started out with 4x 123 cells (approx 14 volts in) and drove a Luxeon III at 700ma until the batteries were almost dead (around Vf of the Lux III - 3.5 volts), would you expect this setup to hit at least 75 % efficiency over the Vin range ?

 

[Kram]

I like this circuit! I've used a variant of it a number of times to provide constant current across a range of voltages. My variant used two NPN transistors instead of an NPN and a FET. The FET would be better than the NPN for targeted currents of, say, more than 200ma (depends on the maximum current rating of the NPN transistor).

It should be noted that the circuit goes out of regulation at the LED Vf + 0.6 + the drop across the FET or NPN. It will continue to work (i.e. pass some current through the LED) all the way down to 0.5 or 0.6V.

It is high efficiency and simple. Actually this is great for 4 cells or for a 9volt battery.

Thanks.

Mark

 

[UK Owl]

[ QUOTE ]
As an example, if I started out with 4x 123 cells (approx 14 volts in) and drove a Luxeon III at 700ma until the batteries were almost dead (around Vf of the Lux III - 3.5 volts), would you expect this setup to hit at least 75 % efficiency over the Vin range ?

[/ QUOTE ]

Unfortunately this is only a slightly cleverer way of limiting current than using a simple resistor. If you want high efficiency you would do far better by using one of the many 'buck' convertor chips on the market. Given the price of 123 cells when used in a high drain situation such as this, an investment in a such a chip would soon pay for itself.


[ QUOTE ]
Actually this is great for 4 cells or for a 9volt battery.

[/ QUOTE ]

I have built several of these built onto a battery snap (solder flying leads from terminals to circuit board & fix with epoxy resin) recycled from a dead PP3 (9v battery), and find the entire circuit & a small switch will fit on a piece of veroboard this small comfortably.

p.s. I used a FET for two reasons.

Firstly because I believed there would be a lower dropout voltage. The ST72A has a 'full on' resistance of about 6 ohms (according to my DMM), at 30 milliamps this means a volts drop of about 0.18 volts, as opposed to the 0.6 volts you would see across a silicon transistor, with a better FET this would be even lower.

The second reason I already mentioned - I had loads of them and just wanted to find a use for some !

Versions I have made (driving white LED's) remain stable down a supply voltage of about 4.1 volts, and are thus perfect for use with either a single PP3 or a bank of 4 AA NiCd / NiMh cells in a battery holder with terminal clips similar to those seen on PP3 batteries.

If you wanted a lower dropout voltage you could always use a much smaller value feedback resistor (R1) e.g. 0.5 ohm, and instead of taking this straight to the base of the transistor amplify the voltage with a rail-to-rail output op-amp, but this greatly increases the number of components needed i.e. an op-amp and a couple more resistors.

This does tend to complicate the circuit a little and removes the novelty of being such a simple circuit.

 

[Kram]

I'm curious to know what kind of small switch you used. I've built similar contraptions, but have had problems finding a switch that is small enough.

Mark

 

[Kram]

I forgot to mention in my earlier posts, but the value of R1 is what determines the current through the LED. The way the circuit works is that the NPN transistor turns "on" when the voltage at the "top" of resistor R1 reaches between 0.55 and 0.65 volts. That causes the voltage at the gate of the FET to drop, thus turning off the current through the FET. Because these two forces essentially balance each other, you have a feedback loop that limits the current to about 0.6V/R1. So, if you want a current of 100ma, the value of R1 would be about 6 ohms (0.6V/0.1A).

A fixed, fairly large value for R2 is OK, as long as it passes enough current to allow the FET to turn "on". With the NPN transistors I've used, I typically use about 10K. You might be able to get by with larger resistance, depending on the gain of the transistor you are using.

Mark

 

[HarryN]

Jack

Why don't you make a list of parts you need for toys. Maybe we can cobble together a parts bin for you.

 

[UK Owl]

[ QUOTE ]
Kram said:
I'm curious to know what kind of small switch you used. I've built similar contraptions, but have had problems finding a switch that is small enough.

Mark

[/ QUOTE ]

They are just some small PCB mount switches that I got when I bought a big bag of mixed components (approx 2 kilo's - mainly various size electrolytic capacitors, but also contained a lot of LED's and a range of assorted transistors and diodes. The bag also contained about 20 of these switches. Unfortunately I have used most of them) for £3 at a computer jumble about 4 or 5 years ago.

Here is a pic of some on circuit boards fitted onto PP3 battery snaps from a previous posting I made.

 

[Kram]

Thank you! Those are cool. In the photos, I can't see that any of your circuits have more than one transistor, but it's difficult to tell because of the blurring effect of the epoxy.

That switch looks very much like the garden variety slide switches that Radio Shack sells. Don't they tend to get clogged with the epoxy around them like that?

Since you refer to a previous thread, I'm now going to have to do a search to find the earlier posts.

Thanks, again.

Mark

 

[reviewum]

Hey,

I think some very specific DIY articles for noobs like me would be GREAT!! I see schematics all over the net, but these aren't very helpful for newbies like me.

The circuits I see discussed / requested the most often seem to be constant current, step up, etc. For example:

Constant Current Circuits: The simple LM317, etc.
Step Up Circuits: the joule thief / satcure / linverter type circuits.

I'd love to see a site that has more than the schematics and one or two very blurry photos. Something similar to what Doug Owen has done (in his excellent teaching ability) by providing actual close-up pictures of the circuits completed in both pre-solder and post-solder stages.

Noobs like me have a hard time following schematics and this is where pictures and good instructions really help out. Of course, like Pablo said in another thread, "an image is better than a 1000 words" so a few nice pictures would be better / easier than instructions.

If someone / a group wants to put things together I can host it.

I love this community spirit!

 

[Doug Owen]

[ QUOTE ]
Kram said:
I'm curious to know what kind of small switch you used. I've built similar contraptions, but have had problems finding a switch that is small enough.



[/ QUOTE ]

Those of us Yanks with a mind to can buy either the small slide switch or the small PB switches shown from Jameco Part number 109170 ($.31) and 119010 ($.27) respectively.

Doug Owen

 

[Jack_Crow]

Harry N and company,

The current project has to do with a low power FM stereo radio station. A bunch of radio clubs back home sent me parts to make this for the troopers.

Guys from the gun rights community are sending me media to play.

One feature I have to add is a 'command receiver'.

Keep in mind this is a military base in a somewhat unhappy country. This base has a PA system for making announcements. I want to give them the ability to make announcemtnes on the radio station. I was going to steel some tricks from the ham radio community.

Purchased used an Icom R2 radio. It has tone decode and I can exploit that signal.

Somewhere in there is a logic change when the tone decoder hears it's signal. I used similar signals when I was building ham repeaters to key the transmitter.

In this case I will use it to activate a relay. That relay will switch the audio paths from the music program to the audio output of the command receiver.

The command transmitter is a UK spec FRS radio. US Spec FRS radios tx in the 462 range. In the UK they have eight channels in the 446 range. Right smack in the middle of the US Ham UHF allocation. In Iraq it will be years before they get around to ham radio, so Im not worried.

So I just ordered a pile of stuff from RadioShack.com. Relays, power mosfets. Sandy from the gun club sent me a bunch of good parts, so most of what I need here.

Also ordered an antenna analyzer. Been a lot of questions about how well antennas are working here, this should end them.

What I need to order next are some adapters for the analyzer, so it will fit Type N , TNC, and BNC.

Here is a link to some photos from this base, toward the end are the ones of the radio station.

That's the news of the moment.

Jack Crow in Iraq

/ubbthreads/images/graemlins/wink2.gif

http://www.obscure-reference.com/hosted/RadioMike/index.html

 

[14C]

Jack, let me know what you need for N, BNC and TNC adapters.
I have a stash from my radar days. Also have some SMAs.