Gen question - which is better practice? 1) Unused leads 2) Diode over 78XX

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Not building anything other than my knowledge base.

1) Which is better practice? To have unused outputs flying? Or to have them tethered to either ground or VSS? (see figure 1) My thought on the matter is even with an exceptionally high resistance, there will still be a current draw. What do you say?

2) To have a diode over a regulator? (see figure 2)
1596232802708.png………. 1596232816999.png
 

jpanhalt

Joined Jan 18, 2008
11,087
Outputs are not the concern. What magic makes you think they are?

As for the regulator, some survive Vout > Vin; some don't. Understand the datasheet.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
What magic makes you think they are?
I don't have an opinion. Just asking if there is a preference. Since I don't know everything - just about anything is worth asking. Nobody learns when they assume sufficient knowledge. But when one questions - one can become the next Einstein. Not that I aspire to that level of knowledge and thinking ability; I just don't want to remain happily stupid.

As for data sheets - some are clear, some are not so clear. Asking those with superior knowledge is a way of learning. What I take from your answer is that you're having a bad day and taking it out on others.
 

MrSoftware

Joined Oct 29, 2013
2,188
Always check the datasheets for the parts. Some have specific instructions for unused pins, others do not. If it's not specified in the datasheet but still not clear, ask the manufacturer or post in their forums. I used a class-d audio amp once and only needed one of the outputs, it was unclear if the other should be left floating or not, I ended up posting in their message forums and eventually got a response from one of their engineers. This was for a life critical safety device so I preferred not to just take a guess at it. But you are correct, sometimes your use case is not covered well in the datasheet leaving the answer unclear.

For the diode question I would also refer to the datasheet, and/or consider the rest of the circuit before deciding.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Part of what drove me to ask is that recently I built a small circuit to attenuate the output of an audio amplifier. I got the impression that leaving an audio amp unloaded was not good practice. As for the diode, I read somewhere that some people are using them in some or in all cases. Not having seen (or heard of) that before I thought I'd get some advice regarding it.
 

dl324

Joined Mar 30, 2015
16,846
1) Which is better practice? To have unused outputs flying? Or to have them tethered to either ground or VSS? (see figure 1) My thought on the matter is even with an exceptionally high resistance, there will still be a current draw. What do you say?
I'd leave unused outputs floating.
2) To have a diode over a regulator? (see figure 2)
The need for that diode depends on the size of the capacitor on the output.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
The need for that diode depends on the size of the capacitor on the output.
From what I was reading I got the impression that if power is cut and there's a large amount of charge on the output filter cap it could blow the regulator. In the many years I've known about electronics I've never heard of a diode bridging back to the input. It makes sense. It might not be necessary, but if I build something with a high capacitance it's good to know what to expect.

With regards 78XX's I've got a whole bunch of 05's and 12's. Don't have plans for them at present. And with PWM, I think I'd be more prone to go that route when I need a particular voltage solution. Or current solution.
 

jpanhalt

Joined Jan 18, 2008
11,087
From what I was reading I got the impression that if power is cut and there's a large amount of charge on the output filter cap it could blow the regulator. In the many years I've known about electronics I've never heard of a diode bridging back to the input.
Unfortunately, your experience seems unusually limited:
Capture.PNG
That's a very old device.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
@jpanhalt yes, "Unfortunately" for me; isn't it?!

I don't know tons. I know a little. Often that's enough to get me into trouble. But at other times I manage to come up with solutions that work. I'm still working out that other post - enter a room, light on, leave the room, light off. I'll update my progress when I have time to tinker with it. If ever.

On another note, when I actually DO know more than someone else, I don't try to make them feel small. I realize that sometimes small people put others down in an effort to make themselves feel bigger. Two answers from you, and both have made me feel smaller. Thank you.
 

OBW0549

Joined Mar 2, 2015
3,566
1) Which is better practice? To have unused outputs flying? Or to have them tethered to either ground or VSS? (see figure 1) My thought on the matter is even with an exceptionally high resistance, there will still be a current draw. What do you say?
Leave unused outputs unconnected.

2) To have a diode over a regulator?
Consult the regulator's data sheet.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Moderators: A thought occurs - we have the "Like" button, and I use it. Sometimes I'd like to thank a person for their answer. It doesn't always rise to the level of liking an answer, but a Thank You button would be, IMO, useful. Please consider.
 

jpanhalt

Joined Jan 18, 2008
11,087
@jpanhalt yes, "Unfortunately" for me; isn't it?!

I don't know tons. I know a little. Often that's enough to get me into trouble. But at other times I manage to come up with solutions that work. I'm still working out that other post - enter a room, light on, leave the room, light off. I'll update my progress when I have time to tinker with it. If ever.

On another note, when I actually DO know more than someone else, I don't try to make them feel small. I realize that sometimes small people put others down in an effort to make themselves feel bigger. Two answers from you, and both have made me feel smaller. Thank you.
My first answer (post #2) was no different than post #10 's answer.
 

Delta Prime

Joined Nov 15, 2019
1,311
Hi there :)always worry about outputs diode D2 is there to protect the regulator damaged of the output. It may have a feedback voltage come back from a load. D2 Is also connected in the reverse bias, it will absorb current spikes to protects this circuit.:)
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
My first answer (post #2) was no different than post #10 's answer.
Though I'm not the one to tell anyone how they should speak, I'll redress what you said in different words:
Outputs are not the concern. What magic makes you think they are?

As for the regulator, some survive Vout > Vin; some don't. Understand the datasheet.
You don't need to be concerned with the outputs. Leaving them disconnected won't hurt anything. The important thing is to not leave inputs floating. As for the regulator, some survive Vout > Vin; some don't. The data sheet should address that issue.

No mention of "Magic" and no need to say "Understand the data sheet" in a way that smacks of an underlaying tone of "Understand the data sheet you moron!"

I'm sure you didn't mean to give the impressions I took. Nevertheless, it's how I take it. Remember, How you say is as important as What you say.

Again, I'm not the one to tell others what to say OR how to say it.
 
Last edited:

AnalogKid

Joined Aug 1, 2013
10,987
1. Digital logic outputs (simple gate chips, complex gate chips, etc.) - leave floating.

Analog output pins (opamps, comparators, multipliers, etc.) - usually they can be left floating. Often they are tied to something because of input pin requirements, like connecting an unused opamp in a package (like 1/4 of an LM324) as a voltage follower with the + input tied to some intermediate voltage.

2. Diode ... The vast majority of the commercial/industrial product schematics I've seen do not have the reverse voltage diode. I've seen it much more often in automotive and MIL applications, but still no more than 75% of the time. For any home project, why not? It's 12 cents on a bad day, or next to nothing if you still have a few hundred on a reel leads left over from a Hamfest purchase in the 80's.

ak
 
1. unused inputs - whatever uses less power
2. The diode. Most cases you don't need it. Any lab supply, you do.
I have a 723 based supply that will kill the 723 if back fed.

You also need to look at your device to see what happens when the device is unpowered, but gets a signal.
The Vcc+0.3 V absolute maximum matters when Vcc is zero or off.
 
Last edited:

ScottWang

Joined Aug 23, 2012
7,397
Where is the input filter capacitor like as 1000 uF or 1500 uF or 2200 uF/25v?

The unused input pins need to connected to Ground and the output pins just leave them floating.

Some people may connected the unused input pins with a 10K to Vcc or ground, but in many industrial application can't do that, because it may catch the noise and causes an error action.
 

crutschow

Joined Mar 14, 2008
34,285
Diode--
The diode is only really required if the input source is rapidly clamped to ground faster than the output load can discharge the output capacitor, giving a reverse bias across the regulator.
If the input is a rectifier-capacitor that feeds only the regulator, and the supply is turned off by removing the AC to the rectifier, then there is not a problem, since the input will collapse no faster than the output, as it's the output load that is draining the input (though the regulator) as well as the output capacitor under that condition.

The lab supply problem is if you connect a voltage source, such as a battery, to the output with the input off.
That would require a diode for protection.
 
Last edited:
Top