7.6 volts from a voltage regulator - Electronic Circuits Projects Diagrams Free

Electric Rain · 25th August 2003, 05:08 AM

Can someone help me figure out how to get 7.6 volts from a FIXED voltage regulator? I don't want to use an adjustable one. But can I like... take the voltage down from an 8 volt reg. with diodes or something? What should I do?

nettron1000 · 25th August 2003, 05:22 AM

Why do you need to get the voltage down to 7.6 ?

stevez · 25th August 2003, 10:41 AM

I've seen situations where the adjustment terminal had a diode inserted with the author indicating that it changed the output by 0.7 volts. Don't know if it works or how regulation is impacted. Author suggested multiple diodes to get other output voltages.

Mosfet · 25th August 2003, 01:18 PM

The short answer, use a 7808.
They are a little harder to find but the extra 0.4 volts shouldn't matter much.
If you need exactly 7.6 volts use a LM317 and dial it in.
Another way is to raise the ground reference pin of a 7805.
You could use a 2.6 volt zener diode.
But unless your current needs are very small the zener would need to be a 2 watt package.
These are expensive and difficult to find.
A high wattage zener can be made by adding a transistor to it.
See my post on pre-regulators.
http://www.electro-tech-online.com/v...ghlight=#13226
Replacing the Darlington pair, Q1 & Q2, with a standard voltage regulator lets you make a regulator at any voltage [within reason].

laroche73 · 25th August 2003, 11:56 PM

I think Nettron 1K asked the right question. Not hard to do with a fixed regulator, but why do you think it's necessary? The problem you're trying to solve may be satistied without such specific voltage requirements.
It's sometimes hard to tell from short posts whether the requests come from a novice who doesn't know any better, or an expert who is looking for a very specific solution. There are clues, though.

I'm just pointing out that posts which provide some background as to the reason "why" often allow the responder to come up with a different (and maybe better) solution.

- CAL

Sebi · 26th August 2003, 04:30 AM

You can do this with 7805 or 7806.

nettron1000 · 26th August 2003, 04:38 AM

Well put Laroche, my thoughts exactly.

Electric Rain · 26th August 2003, 05:24 AM

Quote:

Originally Posted by Sebi

You can do this with 7805 or 7806.

Hey, thanks Sebi! You always seem to answer people’s questions well.

But I'm afraid that I'm new...ish to electronics, so I don't know what values the caps and resistors should be...

ops: Do you think you can help me with those please? :lol:

ops:

Sebi · 26th August 2003, 10:42 AM

All condensers 100nF. With unmarked resistor You can adjust the output voltage. I'm too lazy to calculate it, always made with experiment.(start with 200ohm)
For calculations keep in mind: the 78xx regulators GND pin suck about 2mA (small depend from manufacturer).

Electric Rain · 26th August 2003, 11:47 PM

Quote:

Originally Posted by nettron1000

Why do you need to get the voltage down to 7.6 ?

Why? Well, if you must know, I'm building an interface that lets me use my PSX Pad as a PC Pad. And allot of people say that you can use 9V; however, the rating on the PSX pad is 7.6V. And I might start selling these from my future website. So I don't want to take any chances. (I.G. Someone's pad(s) blowing up :lol: Disclaimer or no.) So anyway, that's why.

And Sebi, thanks! :wink:

laroche73 · 27th August 2003, 03:40 PM

What you do with that 7.6V is your own business, E-rain... I was curious, though. 2cd the good vibes on Sebi, always comes back with a thoughtful answer; Mosfet presented some good ideas, too.

I saw the method Sebi mentioned many years ago, and always liked it for simplicity. It uses the quiescient (kwhy or Kwee-es-scent, what the part consumes on it's own) current flowing through the regulators' ground pin to raise the output voltage by a fixed amount. Check the following links, you'll find this circuit with equations near the end.

http://www.fairchildsemi.com/ds/LM/LM7805.pdf

http://www.national.com/ds/LM/LM78L05.pdf

Keep in mind that Iq is loosely spec'd, ~ 2-8 ma. No problem if you only make a few units and hand-select the resistors, but it's not something you would depend on for higher volume production. The regulators you have on hand may have closely matched Iq's (no pun intended), but then they probably come from the same batch. Always use the worse-case specs from the data sheet when making a production design, and then de-rate by another 10%, if the cost is small.

Here's why I asked for the end application. The first thing I'd do is take apart the PSX pad and see if it has it's own 5V regulator. The numbers mentioned (7.6V - 9V) suggest there's one inside. 7.6V is a typical minimum voltage a designer would spec to account for the insertion loss of a 7805 type regulator (~ 2V). It's kind of redundant to regulate the regulator if that's the case (there are situations where you may want to do this, such as spreading heat dissipation around the various physical components. It's inefficient to do so, and you're better off reducing power dissipation in the first place. Ex: using switching regulators for digital logic).

A quick web search didn't turn up the internals for the PSX pad. Sometimes it's a quicker to figure things out on your own. Don't be afraid to take things apart, even if you can't put them back together. Learning's worth more than stuff. Good luck!

- CAL

Roff · 27th August 2003, 05:56 PM

Sebi's circuit doesn't intentionally use quiescent current (Iq) to raise the output voltage. In fact, he specified a large feedback current (200 ohms suggested, which yields 25ma) to swamp out Iq.
The output voltage will be

Vout=5+100*(5/200+Iq)

If Iq=2ma, then Vout=7.7v

You will need a least 10 or 11 volts on the input to make this work.

laroche73 · 27th August 2003, 09:08 PM

you're right, Ron. perhaps I should stick to digital...

ops:
The feedback current should be >= 5x Iq, according to the datasheet, so the effect of a varying Iq is about 0.1V/mA in this circuit (Vo = 8V @5mA Iq). Iq varies a bit with temp and input voltage.

And you're correct, the input voltage should be ~ 2-3V above the regulator's output voltage for the 78xx series.

My money's still on a regulator already being present in the PSX pad. Anyone care to verify/disprove? And Ron, you can be my proofreader anytime :wink:

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

25th August 2003, 05:08 AM	(permalink)
Electric Rain	7.6 volts from a voltage regulator Can someone help me figure out how to get 7.6 volts from a FIXED voltage regulator? I don't want to use an adjustable one. But can I like... take the voltage down from an 8 volt reg. with diodes or something? What should I do? __________________ When life gives you lemons... make a battery.

25th August 2003, 05:22 AM	(permalink)
nettron1000	Why do you need to get the voltage down to 7.6 ? __________________ Two roads diverged into a woods...i took the one less travelled by. R. Frost.

25th August 2003, 10:41 AM	(permalink)
stevez	I've seen situations where the adjustment terminal had a diode inserted with the author indicating that it changed the output by 0.7 volts. Don't know if it works or how regulation is impacted. Author suggested multiple diodes to get other output voltages. __________________ stevez

25th August 2003, 01:18 PM	(permalink)
Mosfet	The short answer, use a 7808. They are a little harder to find but the extra 0.4 volts shouldn't matter much. If you need exactly 7.6 volts use a LM317 and dial it in. Another way is to raise the ground reference pin of a 7805. You could use a 2.6 volt zener diode. But unless your current needs are very small the zener would need to be a 2 watt package. These are expensive and difficult to find. A high wattage zener can be made by adding a transistor to it. See my post on pre-regulators. http://www.electro-tech-online.com/v...ghlight=#13226 Replacing the Darlington pair, Q1 & Q2, with a standard voltage regulator lets you make a regulator at any voltage [within reason].

25th August 2003, 11:56 PM	(permalink)
laroche73	7.6v ? I think Nettron 1K asked the right question. Not hard to do with a fixed regulator, but why do you think it's necessary? The problem you're trying to solve may be satistied without such specific voltage requirements. It's sometimes hard to tell from short posts whether the requests come from a novice who doesn't know any better, or an expert who is looking for a very specific solution. There are clues, though. I'm just pointing out that posts which provide some background as to the reason "why" often allow the responder to come up with a different (and maybe better) solution. - CAL

26th August 2003, 04:38 AM	(permalink)
nettron1000	Well put Laroche, my thoughts exactly. __________________ Two roads diverged into a woods...i took the one less travelled by. R. Frost.

26th August 2003, 10:42 AM	(permalink)
Sebi	All condensers 100nF. With unmarked resistor You can adjust the output voltage. I'm too lazy to calculate it, always made with experiment.(start with 200ohm) For calculations keep in mind: the 78xx regulators GND pin suck about 2mA (small depend from manufacturer).

27th August 2003, 03:40 PM	(permalink)
laroche73	voltage regulators What you do with that 7.6V is your own business, E-rain... I was curious, though. 2cd the good vibes on Sebi, always comes back with a thoughtful answer; Mosfet presented some good ideas, too. I saw the method Sebi mentioned many years ago, and always liked it for simplicity. It uses the quiescient (kwhy or Kwee-es-scent, what the part consumes on it's own) current flowing through the regulators' ground pin to raise the output voltage by a fixed amount. Check the following links, you'll find this circuit with equations near the end. http://www.fairchildsemi.com/ds/LM/LM7805.pdf http://www.national.com/ds/LM/LM78L05.pdf Keep in mind that Iq is loosely spec'd, ~ 2-8 ma. No problem if you only make a few units and hand-select the resistors, but it's not something you would depend on for higher volume production. The regulators you have on hand may have closely matched Iq's (no pun intended), but then they probably come from the same batch. Always use the worse-case specs from the data sheet when making a production design, and then de-rate by another 10%, if the cost is small. Here's why I asked for the end application. The first thing I'd do is take apart the PSX pad and see if it has it's own 5V regulator. The numbers mentioned (7.6V - 9V) suggest there's one inside. 7.6V is a typical minimum voltage a designer would spec to account for the insertion loss of a 7805 type regulator (~ 2V). It's kind of redundant to regulate the regulator if that's the case (there are situations where you may want to do this, such as spreading heat dissipation around the various physical components. It's inefficient to do so, and you're better off reducing power dissipation in the first place. Ex: using switching regulators for digital logic). A quick web search didn't turn up the internals for the PSX pad. Sometimes it's a quicker to figure things out on your own. Don't be afraid to take things apart, even if you can't put them back together. Learning's worth more than stuff. Good luck! - CAL

27th August 2003, 05:56 PM	(permalink)
Roff	Sebi's circuit doesn't intentionally use quiescent current (Iq) to raise the output voltage. In fact, he specified a large feedback current (200 ohms suggested, which yields 25ma) to swamp out Iq. The output voltage will be Vout=5+100*(5/200+Iq) If Iq=2ma, then Vout=7.7v You will need a least 10 or 11 volts on the input to make this work.

27th August 2003, 09:08 PM	(permalink)
laroche73	you're right, Ron. perhaps I should stick to digital... ops: The feedback current should be >= 5x Iq, according to the datasheet, so the effect of a varying Iq is about 0.1V/mA in this circuit (Vo = 8V @5mA Iq). Iq varies a bit with temp and input voltage. And you're correct, the input voltage should be ~ 2-3V above the regulator's output voltage for the 78xx series. My money's still on a regulator already being present in the PSX pad. Anyone care to verify/disprove? And Ron, you can be my proofreader anytime :wink: