5V or 12V TPS?

It will give a clean 5 volts and is pretty well bomb proof.

Not quite, because the 78xxx series regulators have a tendency to oscillate you use caps before and after which stabilise the regulator and prevents/bypasses the oscillations to ground instead of feeding them into whatever circuit you are powering. Hence bypass cap. You'll find in the original national semiconductor data sheets a full explanation of why you need them.

In message , Harry Bloomfield writes

Actually, the diode I'm referring to isn't to prevent reverse polarity, you strap it across the regulator input and output with the anode on the output, it's to prevent the charged cap on the output dumping its charge back through the regulator and damaging it which could cause the pass transistor to fail short circuit, hence presenting the full input voltage (battery voltage in this case) to whatever you are powering.

Riight. So why put a diode on at all? It's the friggin voltage regulator so there's going to be no reverse polarity.

Decoupling to stop the thing self-oscillating. 0.1uF from each outside pins to -ve does the job. Vital in my book as the self oscillating makes them get a lot hotter than needed. Never bothered with the diode myself.

Mark

Exactly. Depends on the application. I often used them in transmitters and the output had totally different decoupling (bypass) values as it was dealing with high RF fields around 100MHz. Mine also had a 100pF and 1000pF caps as well as a 10uH choke in line with the output to stop RF getting back into the regulator and messing with it's internal reference voltage.

Regarding the reverse polarity diode. It's there to protect the regulator, not the device its powering. The regs go bang instantly with RP. No second chances. Hardly a problem at 40p each, just hassle. As already said though, in a fixed position in a car or inside equipment its not needed.

Mark

Err, we're talking cars?

Well I use them as an LED flasher. That needs different components too. ;-)

Spending a couple of pence to protect a 40p device might seem like good value to some. Especially if it also helps protect the device the regulator is powering.

In message , Mark writes

The diode's just belt and braces and probably not all that necessary on a 5 volt regulator unless your battery/wiring is really shot but it's cheap protection. If it were my design I'd have it rather than not for the cost of a diode.

In message , "Dave Plowman (News)" writes

Exactly the reason I mentioned it in the first place.

You're the one asking "WTF is that?". Take a guess who it is that doesn't know what the f*ck they're on about.

In article , David Taylor says...>

Well I never had to bother with them when I did my BTEC HNC Electronics Engineering. Perhaps you'd like to tell me what your qualifications and experience are?

Thought not.

Just to clarify; I was talking more about their use in electronics soldred onto a board as part of something that derives its power from a mains PSU, so impossible to RP the thing once built.

Mark

In article , Clint Sharp says...>

Care to cite the source for that? That's a proper source, not just some post in a forum or newsgroup.

You've done up to HNC in electronics and have never heard of capacitors used for decoupling/bypass?

Pull the other one.

Mark

How about 40 years of personal experience and 25 years using these regulators since they first appeared. Trust me, this aint a pedants type answer (like Plowman et al), its a practical fact. Look at the output on a scope and 8 times out of ten you can see them oscillating away. Without stopping that with a cap, any electronics it's powering will tend to be unstable, although obviously this won't affect powering things like lamps and motors, but as I said earlier, they run hotter than needed when oscillating.

Mark

In article , Mark=20 says...>=20

Oscillating by how much? I built an oscilloscope calibrator using one=20 once. I think I'd have noticed if it was. A few millivolts here and=20 there isn't going to make much difference ITRW. Capacitors wouldn't=20 stop the regulator oscillating - they'd just smooth the output, surely?

The only results I can find mentioning oscillation involve using them=20 in a circuit with high frequencies. You're not going to get that in a=20 car.

FYI, From National Semiconductor:

By Chester Simpson, Member of Technical Staff, Power Supply Design=20 Group=20

Linear regulators

A dangerous precedent was established by the first linear regulator=20 semiconductors sold commercially like the LM7805 type devices: they=20 require no input or output capacitor and are completely stable under=20 virtually any operating conditions. Some of the newer LDO regulators=20 require careful attention to external capacitors to operate in a stable=20 mode. The main reason the 7805 is unconditionally stable is because the=20 power pass transistor was made up of an NPN Darlington as shown in=20 Figure 4. They are usually referred to as ?NPN regulators? for this=20 reason, and are comprised of an error amplifier, voltage reference, and=20 NPN Darlington power transistor.

--=20 Conor.

In article , Mark says...>

Yes.

I said it wasn't needed in this situation.

I'm fairly sure I already have. BEng Computing & Electronics.

Or just read a bloody text book.

In message , Conor writes

Page 21.

explains why linear regulators oscillate and what you can do to stop it happening.

Enough?