Hello everyone,
So I have this really strange problem with a power supply, whenever it is physically near a certain frequency the voltage collapses. For example, I make a 555 timer circuit on a breadboard, I can supply 5 volts to it without a problem from the supply. However when I then connect a inverter ic to the output of the timer and try to power it, the supply voltage drops to 1 volt. I tried adding a capacitor on the output of the supply also a flyback diode but that doesn't make a difference. Now the very wierd part is, when i connect the circuit to a 12v pc power supply the circuit is able to run just fine but when that very circuit is near my original power supply the ouput still collapses (even though it isn't connected in any way). For me it is like magic, does anyone have a explanation for this? This also happens with low frequencys like 300hz.

Note that my 'power supply' essentially is just a 500 watt pc power supply with a buck boost converter and a voltage/current display hooked up to it. I used this buck boost converter, https://nl.aliexpress.com/item/LTC3...020.html?spm=a2g0s.9042311.0.0.27424c4dSZY0wO.

 

Can you show your schematic?

 

Can you show your schematic?

The schematic of the 555 timer or power supply?

 

This is one of the circuits that will cause the problem, it works perfectly without the 74hc4049n ic. Note that this isn't the only circuit that messes up my psu.

 

You need to post the schematic of the 555 timer and how it is connected to the inverter IC (Including part number of the inverter IC) Also in block form show the PC power supply feeding the buck converter. also make it clear if it is the output of the PC power supply that drops to one volt or is it the output of the buck converter that drops to 1 volt while the voltage from the PC power supply remains constant. (You don't say which voltage output from the PC power supply you are using to feed the buck converter.) You also do not say weather you added the capacitor to the output of the PC power supply or to the output of the buck converter. (What was the value of the capacitor ?)

Les.

 

I hooked up capacitors to the boost/buck converters output (since pc psu's already have big filtering caps on their output) I tried 100-6000uF electrolitic capacitors and/with 100nF ceramic ones. Still I don't think the problem lies here since the psu also starts acting crazy when a circuit that generates a frequency and is not hooked up to it is near it (within a meter). The pc power supply stays at a steady 12v output, the boost/buck converter drops. The boost/buck will work with a 5-32v input and i'm suppling 12v. This is the project I recreated to make my psu:https://www.instructables.com/id/Build-your-own-Variable-Lab-Bench-Power-Supply/

 

In your first post when you say "Now the very wierd part is, when i connect the circuit to a 12v pc power supply the circuit is able to run just fine but when that very circuit is near my original power supply the ouput still collapses" by circuit do you mean just the and inverter or do you mean that you are using a different 12 volt power supply to feed the input of the buck regulator ? My line of thinking is that radiated interference from the 555 nad inverter is getting into the long wires between the buck regulator and potentiometers. With the inverter connected to the output of the 555 the edges of the square wave on the output of the inverter will be faster than the edges at the output of the 555 so they will contain higher frequency components which will radiate more than the slow edges. This radiated signal may be being picked up on the wires to the potentiometers.

Les.

 

I am guessing that the "500 watt PC power supply" is a personal computer power supply. I would not expect to be able to use one of those for any other application. They have very specific current draw requirements and a minimum load requirement on the 5 volt line.. Also there is a limit on the total watts drawn, so if your load on the 5 volt line takes things close to that power limit, and then yur load on the 12 volt line takes you over that limit, the output will indeed collapse.

 

In your first post when you say "Now the very wierd part is, when i connect the circuit to a 12v pc power supply the circuit is able to run just fine but when that very circuit is near my original power supply the ouput still collapses" by circuit do you mean just the and inverter or do you mean that you are using a different 12 volt power supply to feed the input of the buck regulator ? My line of thinking is that radiated interference from the 555 nad inverter is getting into the long wires between the buck regulator and potentiometers. With the inverter connected to the output of the 555 the edges of the square wave on the output of the inverter will be faster than the edges at the output of the 555 so they will contain higher frequency components which will radiate more than the slow edges. This radiated signal may be being picked up on the wires to the potentiometers.

Les.

Yes I indeed mean the 555/inverter circuit, however I don't use another psu for the boost/buck but I just hook the 555/inverter directly to a 12v psu without a converter. Still the boost/buck voltage collapses, i'm thinking that the signal is maybe getting to the boost/buck trough the mains earth. I tried to illustrate my problem a bit because im not a native speaker and maybe my explenation isn't 100% clear. I hope this picture helps:

Also, if signals are reaching the pots, how do i prevent this from happening?

 

Try connecting the pots with very short wires to the buck converter or removing the pots and re fit the original preset pots to the buck converter. If this clears the problem then using screened cable to the pots may work.

Les.

 

My guess is something inside the PSU is picking up the EMR from the inverter circuit. If you're using a digital PSU, try using an analog one, or try to shorten all wires on your inverter, and make sure none of the wires have any loops. Any inductors could be part of the issue too.

I had something similar when working on a project that involved radios (Motorola radios, like the police use). I had one radio sitting on top of my bench power supply, and I could not figure out my circuit would loose power every time I keyed the mic on the other radio (it was Bluetooth connected to one of the radios). By chance I happened to be looking at my power supply at the same time I keyed the mic and I saw the voltage quickly scroll down as I keyed the mic (it was a digital PSU), and scroll back up as I released the mic. I moved the radio away from the PSU and suddenly all was well. Apparently the radiation from the radio was causing an issue with some component inside the PSU, causing it to turn itself down. Maybe you're seeing something similar.

 

Certainly power supplies with inadequate filtering on the outputs and inputs will suffer from any signals on the external connections.

 

Try connecting the pots with very short wires to the buck converter or removing the pots and re fit the original preset pots to the buck converter. If this clears the problem then using screened cable to the pots may work.

Les.

I'm actually using the original pots, I only extendend the wires to the converter. Back when I made this psu I could not find any potentiometers appropriate for the converter. However in the past year I came along more and more chinese sellers so I'm sure I can score a better potentiometer. I don't know if this makes any difference but the pots I now use are trimming pots (the ones you need to adjust with a screwdriver), would it help if I replace them with the big potentiometers? I came across these ones. Somehow 200k potentiometers are hard to find...

 

I'm actually using the original pots, I only extendend the wires to the converter. Back when I made this psu I could not find any potentiometers appropriate for the converter. However in the past year I came along more and more chinese sellers so I'm sure I can score a better potentiometer. I don't know if this makes any difference but the pots I now use are trimming pots (the ones you need to adjust with a screwdriver), would it help if I replace them with the big potentiometers? I came across these ones. Somehow 200k potentiometers are hard to find...

The main issue will be the length of the wires. Try putting them back on the board to see if the problem goes away, then try using shielded cable for the connection, audio cable will do.

 

As you are using the original preset pots still on the buck converter the idea of electrical noise pickup at the voltage or current comparitor inputs on the buck converter is no longer a valid theory. I suggest powering the buck converter from a 12 volt battery (Such as an SLA alarm battery or a car battery.) Use a fuse (Say 2 amps.) between the battery and the buck converter. this will prove if the problem is with the buck converter ot the PC power supply. Another thought as occured to me. Is the voltage actually going down to 1 volt or is it just the voltage / current display that is indicating that is going down to 1 volts.
EDIT. I have just noticed kubeek's post #14. When you said "I only extendend the wires to the converter." I assumed that you meant the power input and power output wires but as you did not say which wires then you could have meant that you have removed the presets from the buck converter board and put them at the end of extended wiring. If this is the case we are back to radiated electrical noise being picked up on these wires.

Les.

 

Your '4049 IC in the circuit shown has several unconnected (floating) inputs. That is a no-no for CMOS circuits, as it leads to unpreditable behaviour. The IC is probably oscillating merrily and the EMR is affecting the power supply.

 

Your '4049 IC in the circuit shown has several unconnected (floating) inputs. That is a no-no for CMOS circuits, as it leads to unpreditable behaviour. The IC is probably oscillating merrily and the EMR is affecting the power supply.

Good eye!

 

Your '4049 IC in the circuit shown has several unconnected (floating) inputs. That is a no-no for CMOS circuits, as it leads to unpreditable behaviour. The IC is probably oscillating merrily and the EMR is affecting the power supply.

This could indeed be a reason, can I just connect every unconnected input to ground or vcc to solve this problem? I will try that and give an update soon.

 

As you are using the original preset pots still on the buck converter the idea of electrical noise pickup at the voltage or current comparitor inputs on the buck converter is no longer a valid theory. I suggest powering the buck converter from a 12 volt battery (Such as an SLA alarm battery or a car battery.) Use a fuse (Say 2 amps.) between the battery and the buck converter. this will prove if the problem is with the buck converter ot the PC power supply. Another thought as occured to me. Is the voltage actually going down to 1 volt or is it just the voltage / current display that is indicating that is going down to 1 volts.
EDIT. I have just noticed kubeek's post #14. When you said "I only extendend the wires to the converter." I assumed that you meant the power input and power output wires but as you did not say which wires then you could have meant that you have removed the presets from the buck converter board and put them at the end of extended wiring. If this is the case we are back to radiated electrical noise being picked up on these wires.

Les.

I'm 100% sure the converter is the problem and not the psu. The possibility that the extended wires (between the pots and converter) are picking up EMFs and post #16 from alec are the main problems I think. I still think its wierd that even frequencies around 200-400hz also interfere with the converter.

 

A square wave with fast edges has components that go up to a much higher frequency than the base frequency of the square wave. Google "fourier analysis of square wave" and you will find it contains components of all the odd harmonics. The wires to the potentiometers will be relatively high impedence and will pick up interference easily. Try putting the presets back on the PCB or connect them with VERY short wires.

Les.