|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
17th Dec 2017, 5:28 pm | #1 |
Guest
Posts: n/a
|
78xx regulators, reverse protection.
Simple question, do the 78xx series of fixed voltage regulators need reverse current protectoin (a diode twixt input and output) i.e. the load still has volts and the source doesn't. The LM317 datasheets say they do, but I can't find anything in the numerous 78xx datasheets I have looked at.
The reason for asking is that I have killed one (a 7815) which fed a float charged battery, and don't want to go tracking down a non existent fault. |
17th Dec 2017, 6:11 pm | #2 |
Rest in Peace
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
|
Re: 78xx regulators, reverse protection.
I've seen that done in reputable designs that featured 78XX regulators; from personal experience of what can happen if said diode is not fitted, cannot say. However, if the LM317 datasheets state that fitting that diode is a good idea, that's good enough for me! After all, something like a 1N400X is cheap & saves the hassle of a subsequent repair.
Al. Last edited by Skywave; 17th Dec 2017 at 6:16 pm. Reason: General re-write. |
17th Dec 2017, 6:14 pm | #3 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,947
|
Re: 78xx regulators, reverse protection.
I agree. I've got away with it in the past, but protection diodes are good design practice and cost very little. It's odd that the protection diodes can't be included in the regulator silicon though.
|
17th Dec 2017, 6:44 pm | #4 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,208
|
Re: 78xx regulators, reverse protection.
I've seen plenty of commercial designs using 78xx's with the diode between output and input for protection. So I assume it was a good idea.
I normally fit one (1N4007's are cheap) even if it's not strictly necessary. It's not going to do any harm. |
17th Dec 2017, 6:49 pm | #5 | ||
Guest
Posts: n/a
|
Re: 78xx regulators, reverse protection.
Quote:
Quote:
|
||
17th Dec 2017, 7:05 pm | #6 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: 78xx regulators, reverse protection.
Do put them on. During power failure scenarios any large capacitors on the output will discharge into it and bang. The reason they don’t exist inside the package is because it makes them slightly less flexible. If you are using an adjustable regulator they don’t even make sense as both of the output terminals aren’t available on the package itself. The diode across the input/output also breaks a sneaky power boosting function where you stick a low ohm high power resistor across it to knock out some more output current.
Ergo it’s more flexible if they aren’t included. Linear’s LT1084 series does have them though. I’ve tried to blow them up and they just won’t! Edit: to be clear you can use a 78xx regulator as an adjustable one. The regulator just tries to maintain say 5v between out and common. If you stick a divider on it, the lowly 7805 does the same job as an LM317 but cheaper. |
17th Dec 2017, 7:27 pm | #7 |
Dekatron
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 5,271
|
Re: 78xx regulators, reverse protection.
A bypass diode is recommended but for different reasons
__________________
Kevin |
17th Dec 2017, 7:35 pm | #8 | |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,998
|
Re: 78xx regulators, reverse protection.
Quote:
[You always go to great efforts to provide stable HT to a valve VFO - but wandering heater voltage can be just as annoying a cause of frequency-shifts] |
|
17th Dec 2017, 7:39 pm | #9 |
Hexode
Join Date: Apr 2008
Location: Ramsgate, Kent, UK.
Posts: 252
|
Re: 78xx regulators, reverse protection.
It clearly states in the application notes/hints section of the data sheet for both TI and NS versions of the LM78xx or LM340 voltage regulators that the device can be destroyed by either (a) shorting the input or (b) raising the output voltage above the input voltage - e.g. by having a battery as the load - Unless an anti-parallel diode is fitted across the output/input terminals.
"Shorting the Regulator Input: When using large capacitors at the output of these regulators, a protection diode connected input to output (Figure 1) may be required if the input is shorted to ground. Without the protection diode, an input short will cause the input to rapidly approach ground potential, while the output remains near the initial VOUTbecause of the stored charge in the large output capacitor. The capacitor will then discharge through a large internal input to output diode and parasitic transistors. If the energy released by the capacitor is large enough, this diode, low current metal and the regulator will be destroyed. The fast diode in Figure 1 will shunt most of the capacitors discharge current around the regulator. Generally no protection diode is required for values of output capacitance <= 10 μF. Raising the Output Voltage above the Input Voltage: Since the output of the device does not sink current, forcing the output high can cause damage to internal low current paths in a manner similar to that just described in the “Shorting the Regulator Input” section." A 1N4001/2 diode is adequate for such a task, connect cathode (the banded end) to the regulator input for a positive (LM78xx) regulator, or to the regulator output for a negative (LM79xx) regulator. See Kevin's Post #7 for connection diagram. Last edited by RF Burn; 17th Dec 2017 at 7:48 pm. Reason: Added reference to connection diagram |
17th Dec 2017, 7:40 pm | #10 |
Heptode
Join Date: Dec 2014
Location: Lincoln, Lincolnshire, UK.
Posts: 583
|
Re: 78xx regulators, reverse protection.
I almost always use them. I also very often use a fixed regulator at a different voltage as described above. One useful circuit I have come across is the one using a transistor, resistor and capacitor in the feedback to provide ramp up of voltage. Ideal to limit switch on surge with a valve heater etc.
|
17th Dec 2017, 7:58 pm | #11 |
Guest
Posts: n/a
|
Re: 78xx regulators, reverse protection.
RF Burn, thank you for the information, despite looking I couldn't find it. Good practice is fine but without finding out why a bit vague. In my application I have a 7815 with a few diodes in the common terminal to ground to raise the output to 21V (ish) for a 9 cell lead acid battery, does the temperature compensation too (lucky me).
New regulator and a diode this time! (I have some 60V 5A Schottkys in stock). |
17th Dec 2017, 10:02 pm | #12 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: 78xx regulators, reverse protection.
I have never been fond of the idea of shunting charge and energy as a solution to protecting devices though it's common in other circuits such as the usual diode across inductors like relays etc.
I think for the LM78xx regulations it is better to use small input & output capacitors, a 2.2uF tant to 4.7uF tant with 0.1uF caps across them is normally fine to keep the regulator stable and they don't store too much energy. Then have both a series input diode which prevents the regulator being exposed to reverse polarity (especially if it runs off various wall warts and cords), and a series output diode, so a battery as a load for example doesn't discharge back into the regulator. Schottky diodes for lower voltage drops than 1N4001 and help and if required the output of the regulator can be jacked up with a diode in series with the regulator's earth leg. I bypass those with a 0.1uF. If an isolated supply like a wallwart is used, a small bridge rectifier at the input guarantees it will work regardless of the input polarity presented to it. This type of problem has also cropped with attempts to use "ideal rectifiers" (which have a control IC controlling a series pass mosfets)can trying to use them in some battery charge circuits because the battery voltage remains when the regulator powers down. A series diode always solves the problem, but if the currents are high it's wasteful of energy and the efficiencies of the ideal diode can be lost. I built a supply for battery valve radio once that had 45 , 90 and 135V outputs. If I removed the plug which supplied the 135V HT to the radio and quickly plugged it into the 45V connection, the stored energy in the radio's filter caps, destroyed the regulator, so again series output diodes were helpful to prevent this. |
17th Dec 2017, 11:37 pm | #13 |
Dekatron
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
|
Re: 78xx regulators, reverse protection.
Shunting energy away via diodes only works if there is some resistance to limit the current and safely dissipate that energy (for instance, the resistance of the relay coil with a diode across it). Shunting the output of the regulator IC back to the input relies on the combined resistances of the regulator pass transistor and the circuit itself.
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments. |
17th Dec 2017, 11:59 pm | #14 |
Dekatron
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,553
|
Re: 78xx regulators, reverse protection.
I have been trying to get a data sheet for the charge regulator in an e-cig.
The battery is a two terminal device with a circuit that regulates charge when a voltage above the battery voltage is applied. When in use pressing a switch supplies about three amps to the heater connected across the same two terminals. They have a FET with a very low on resistance external to the chip. It would be interesting to fully reverse engineer one. |
18th Dec 2017, 12:17 pm | #15 |
Pentode
Join Date: Nov 2015
Location: Maidstone, Kent, UK.
Posts: 131
|
Re: 78xx regulators, reverse protection.
R F Burn has said it all. Most data sheets refer to this problem and recommend the fitting of the protection diode. The low cost and ready availability of the 1N400n & 1N54-- series make this precaution an easy option as the single pulse performance will cope with most fault situations.
|
18th Dec 2017, 1:00 pm | #16 |
Hexode
Join Date: Apr 2008
Location: Ramsgate, Kent, UK.
Posts: 252
|
Re: 78xx regulators, reverse protection.
For those who worry about 'charge shunting', don't worry because it is an accepted method of protecting devices against back EMF or other reverse voltage situations by design engineers not only in consumer electronics, but in industrial, medical and military applications where reliability is a primary design target.
Omitting the protection diode and trying to keep the capacitance across the output of the regulator to below the recommended maximum of 10 μF to avoid the problem is a dodgy solution at best, since this capacitance will include all the decoupling components... |
18th Dec 2017, 1:55 pm | #17 | |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,082
|
Re: 78xx regulators, reverse protection.
I tend to use inverse bypass diodes as a matter of course, only leaving out if the regulator IC is stabilising an internal rail, from an internal unregulated supply. Only then can I be sure that the output is never greater than the input (and even then, when fault-finding, there's the possibility of a screwdriver crowbarring the unregulated supply... All the regulated rail's decoupling capacitors discharge backwards through the regulator. Not a nice thought.)
There is always the possibility of the diode suddenly failing short-circuit, and bypassing the regulator. But I reckon it's remote. More likely for the regulator itself to fail S/C. Quote:
This does mean that the LM78xx regulators give the correct output voltage right down to zero load, whereas the LM317 needs a minimum load of a few mA. But that's usually not a problem. |
|
18th Dec 2017, 4:14 pm | #18 | |
Guest
Posts: n/a
|
Re: 78xx regulators, reverse protection.
Quote:
Update, new 7815 fitted with both reverse diode and a feed diode (good idea) and all is working well. Thank you all for the advice and comments, in 40 odd years (50 if you include playing as a kid) I have never come across this for 78xx regulators, now I know. Never killed one either! |
|