Simple DC Voltage Regulation Circuit Question

[parabuzzle]

Hello,

I have a little Icom T90a handheld ham radio. I want to build a DC car charger plug for it. (I can't see paying $14+ for something I can build from my parts bin).

Anyway, the manual says:

 * Power supply range is between 5.5–11.5 V DC. NEVER CONNECT OVER 11.5 V DC directly into the DC power jack of the transceiver.
 * BE SURE to use the CP-19 when connecting a regulated 12 V DC power supply.
 * If a battery pack is attached, the voltage of the external power supply must be within 11.5–16 V DC, otherwise, the battery power may be used for operation.

Which almost sounds counter intuitive to me really.. I have discovered if you go over the magic 11.5V the radio has some logic based protection that will allow it to charge but when you attempt to transmit it will display a high voltage error on the screen.

So the parameters are:
 1) must be able to supply between 7.9V - 11.5V (7.8V is the battery when fully charged so if the power is less than that, the battery will be used for transmit)
 2) must not exceed 11.5V

Here is the circuit I designed and have stuck on the breadboard:

http://www.circuitlab.com/circuit/hymzxw/screenshot/1024x768


I have attached the plot of my simulated voltage sweep from 10V to 15V

Now, I have 2 questions:

1) Is that diode needed? I put it in as a protection but I don't know if its just a needless component.
2) Any suggestions?

[Galaxyrise]

How much current does this need to supply to charge the battery?  I think what you've drawn out there is going to have problems delivering much.  At 100mA, the output will have dropped to 9.3V, and your resistor might be really hot depending on what kind you've chosen.

Presuming that's a LM7812, the datasheet specifies a 2V dropout voltage, which means your circuit only works at 14V.  So don't use it when the car is off.

I don't think your circuit simulator gave the diode a voltage drop, which a real diode will have.

My intuition says there should be an inductor on the cigarette lighter side.

[parabuzzle]

The charger draws about 143mA

The Receiver (turned on) with charger draws about 156mA

The transmit draws 0.82A in 1/5 watt transmit mode and 1.4A in 5 watt transmit mode.

All are measured using my Amprobe AM-570 meter.

-Mike

[parabuzzle]

Well, when the circuit is lower than 14V just see the 2V dropout on the input..

ie - 12V in results in 10V out

It seems to resemble a simple voltage divider.

Is it bad to run the regulator like that?

-Mike

[Clint]

I would use a 7809 with a resistor on the GND pin to get the exact voltage you want 11.4v.

[pinkysbrein]

Why not just mod a cheap as chips MC34063 USB car charger?

[Brainless]

The voltage on the cigarette lighter can actually peak to 100v or even more.
You would need protection against that.

[AG6QR]

My first instict would be to use an LM317 with the right pair of resistors to give 11.4V; that's what it was designed for.  But as Clint points out, a 7809 can be used in the same way (with different resistor choices, of course), and I might choose that solution if I had a pile of 7809s and no 317s.  But both of these solutions have the problem of dropout.  When the engine is off, a car's electrical system is pretty close to the desired 11.4V output.  (Note that the original LM7812 solution has a worse problem of dropout).  So a low dropout solution is probably better.  An LT3080?


I went and read the Icom T90a manual, to get the quotes in context.  Where it says "If a battery pack is attached, the voltage of the external power supply must be within 11.5–16 V DC, otherwise, the battery power may be used for operation", the word "external" seems to me to be referring to the input to the CP-19 voltage regulator, not to its output.  So it looks like what you want is to get the input to the radio (output of your regulator) as close to 11.5 as practical without exceeding it.

I would NOT use a resistor divider on the output of the voltage regulator.  Besides wasting power, it causes the output voltage to vary with current draw.

The suggested mc3063 might work, but for radio applications, I'd be a bit wary of switching noise.  I don't know if it would be a problem or not, but a linear solution is simpler (fewer external components) and more likely to be noise-free.

[parabuzzle]

So I just wired this up on the breadboard and it is much more stable:

[SeanB]

I would not bother with the resistor on the 7809, but suggest putting the diode on the power input, and make C1 1000uF 25V and C2 47uF 25V. R1 short circuit and R2 2k2 with a LED in series as power indicator so you are confident there is power.

Take any cheap OHL car charger ( you only want the case and the curly lead so the actual phone it was meant for does not matter) and use the case and lead to hold the 6 components. A small scrap of aluminium as a heatsink for the regulator will help, and just fit it all inside the case after gutting the old stuff out.

I did a similar thing but reused the 34063 SMPS and adjusted it to give me a 9V output by soldering a 3k3 IIRC resistor across the one feedback resistor to up the voltage from 5V to 9V. Slightly better caps than the original Mystrlee ones in there ( got to be poor if there is only a value and voltage with no manufacturer name or logo) and it works well for a year now.

[Gall]

I suggest using LM317 just because they have better specs. 78xx are way too old and generally not recommended for new designs.

[Rick Law]

How about using an LM2577 to boost it first, and then use something like an LM2596 to buck it back down to the exact voltage you want.

With the 2577, you can use power source from 3v to 27v.

[Clint]

So I just wired this up on the breadboard and it is much more stable:

Perfect, cheap simple and effective, also it will last YEARS !