Would a 5 vdc circuit board start to fail if powered with 4.7 vdc too long ?

[Baker in Virginia]

Good Evening ,

Just wondering . We have a circuit board on a laser , that we are having low voltage issues with . I took notice that the board had only 4.7 vdc on it last week and it was acting funny , resetting itself , etc. If a 5 vdc board would have only
4.7 vdc for a long time like a year or two , could components start to fail ?

I raised the voltage to 5.2 vdc and it stabilized . But it faulted out again today .

Thanks so much in advance .

 

[OkiePC]

I've seen similar symptoms with an electronic scale. Anything less than 4.8 volts would make it hiccup. Restoring the full voltage cleared up the symptoms to the best of my recollection. I don't know if there is a risk of damage due to low voltage for a long period of time. Maybe one of the electronics gurus can chime in.

 

[parky]

In my early years, I worked exclusively on electronic circuits mainly logic & the early microprocessors, In my experience it is unlikely to damage components but over voltage can definitely damage electronic components. Digital equipment that works at 5v can fault at anything above 5.6v, I have never seen a fault occur on low voltage, It could be that the low voltage is due to the amount of current drawn is above the spec of the power supply and increasing it makes it worse. It is possible that a component is already failing causing the voltage to drop. I would check the supply off load first, if it drops on load (assuming it is regulated) then it is highly likely that a component is failing. Also I have seen where the smoothing capacitors in the PSU have deteriorated and cause ripple on the supply, this gives the appearance of low voltage. A good example of this was one I was called to a photocell for build back seemed to be working correctly as far as the signal to the PLC, but, while monitoring the plc code the buildback timer kept resetting, my first thoughts where a duplicate timer or bit, however I noticed that the LEDs on the input cards were a little dim, a check with a meter showed only 22 volts but not significantly low to cause a problem, a scope check showed a large ripple, so the ripple negative wave was falling below the level to reliably trigger the input in essence, the input was switching on/off at 100 hz and although on this early plc the program would not see such a fast pulse occasionally it would and reset the timer.

 

[drbitboy]

... a check [of a power supply] with a meter showed only 22 volts but not significantly low to cause a problem, a scope check showed a large ripple, so the ripple negative wave was falling below the level to reliably trigger the input...

Wow, for anyone except parky and the several dozen or so people who read and remember this, the next time this happens to most people it will be "fool me once ...," i.e. the first time. It sounds like a scope in this industry is like an MRI in health care: too cumbersome, usually, to bring along or prescribe, but definitive when all else fails. There are apps that could convert a phone or tablet into poor-man's scope.


[Please excuse the off-topic nature of this comment, but this is what information like this makes me think about]

 

[Aabeck]

I had a problem with a 5V drive, that if the power dropped the CPU would have problems reading the memory, declare the memory corrupt and wire it clean.

Putting in its own 5V power supply fixed the problem.

 

[parky]

you must remember I was once in the pure electronics industry before a computer was less than £1m and a scope was essential, Scopes nowadays are cheap compared when I purchased my first one, I now have a small one it only cost £100.00 and it is also a meter It just happened I had it with me on site. It has also come in handy when we had a problem with an E&H flowmeter, we were getting inconsistent additions to a batch process and again trying to figure out what was the problem I put a scope on the digital output (the flowrate seemed ok) & the waveform was varying in size and misshaped thus missed pulses in the PLC, fortunately, I had ordered a couple of sets of spare cards as the FM's were over £3k each & we had over 20 of them. replacing the I/O card cured the problem.

 

[TheWaterboy]

I'm with parky, likely ripple is the issue here either due to aged caps (common and inevitable) or overloaded PS. You might be able to measure it with a quality meter on the AC setting if the ripple is around 60 hz or so severe that it drops below zero for part of its cycle... but a scope reveals everything.

 

[Lare]

Wow, for anyone except parky and the several dozen or so people who read and remember this, the next time this happens to most people it will be "fool me once ...," i.e. the first time. It sounds like a scope in this industry is like an MRI in health care: too cumbersome, usually, to bring along or prescribe, but definitive when all else fails. There are apps that could convert a phone or tablet into poor-man's scope.


[Please excuse the off-topic nature of this comment, but this is what information like this makes me think about]

Even poorer


Some LCD type multimeters smoothes DC signal, so fast up/down going signal is showed as steady voltage.


Old multimeters used moving coil type panel meter. If signal is fast but slow enought, it maybe shows better how DC voltage is jittering because of old caps.



https://www.youtube.com/watch?v=WzvHYcNhxmY

 

[Jasondelane]

Even poorer


Some LCD type multimeters smoothes DC signal, so fast up/down going signal is showed as steady voltage.


Old multimeters used moving coil type panel meter. If signal is fast but slow enought, it maybe shows better how DC voltage is jittering because of old caps.



https://www.youtube.com/watch?v=WzvHYcNhxmY

That's why God invented min/max

 

[TheWaterboy]

... and then there was Peak.. and he saw it was good...

Min/Max still might not catch a high freq ripple if all readings are RMS.

 

[rajy2r]

I now have a small one it only cost £100.00 and it is also a meter It just happened I had it with me on site.

Could you mention the make and model of the meter. I have been hoping to pick something handy as I have run into ripple issues a bunch of times.

 

[parky]

I looked for it and cannot find it, I believe I loaned it to a friend (Doh!) I ordered it from amazon but I cannot find it on there now and cannot remember the make or model, but there are plenty others, I chose it due to the reviews wasn't expecting much as it was from china, but it worked well, however it's bandwidth was only 20mhz My other scope is too big for traveling with but that one is 200mhz.

 

[drbitboy]

heh

This might work for a portable, in-pocket or in-toolbox, solution for almost all cases.

Here* is an example of someone using that or summat similar. The values 500us and 50us show up on the screen as the abscissa zoom is varied, suggesting sampling is of order 20kHz**. That should be enough to pick up 60Hz noise, or look at what is happening with a PLC running ms-level scan periods.

One would have to fiddle with the resistors on a case by case basis, have multiple cables for typical voltage ranges, or build a box with a rotary switch.
There appears to be some ringing in the data for the spikes shown in that video; that ringing could be in the signal itself, or in the A/D converter system.

The first link recommends GoldWave for the software; I cannot tell if that video uses GoldWave. I have used it for audio in a soundbooth on a PC before and it is solid enough, but I don't know anything about the Android or iOS versions. It has a free trial version that never expires so the risk is low; a lifetime license is all of $45.

* That URL starts near the end of that youtube video; hit pause about 5-7s after it starts for a decent view of the screen.

** i.e. the upper frequency limit for audible sound (for someone with younger ears than mine).