Is MW LRS 100-24 a good power supply? + possible leakage solution question

[DaveK]

Hi,
i'm looking for a low leakage, 24V atleast 3A power supply for my TS100 station.
The Mean Well 100W 24V one is pretty cheap and fits within my budget nicely, i would like to know if it uses any Class Y cap to connect the DC rails between each other.
I know it has an earth pin and definitely atleast 2 caps that are referenced to mains earth, but i am not sure about the third Class Y cap from the pics. I know the 150W version has 1 pair of those that couple high and low voltage negative DC (will cause a bit of leak afaik). If there is a Class Y cap that couples the rails, would removing it cause severe noise in the power lines or the output? I wasn't able to find any PCB pics of the particular model i'm interested in, so if anyone owns one please post pictures of the PCB.
Cheers
Dave

 

[BobK]

What kind of leakage are you talking about?

And it almost certainly has a cap across the power outputs, pretty much all DC supplies do. Why is this a concern to you? There will be a large AC component if you remove the capacitor.

Bob

 

[DaveK]

What kind of leakage are you talking about?

And it almost certainly has a cap across the power outputs, pretty much all DC supplies do. Why is this a concern to you? There will be a large AC component if you remove the capacitor.

Bob

I mean AC leakage that occurs on switching power supplies. It's because of the EMI filters. For example, if i take my laptop charger and set my MM to 20VAC, i can measure 9.78VAC between positive of the supply and mains earth. It's not a big deal for most applications but for something like a soldering iron, i can't have AC leaking from the tip when working on sensitive semiconductors. I obviously have to ground the tip, but the heater is in series with thermocouple and if the leakage from the power supply is too high, it's going to cause high current through the thermocouple and confuse the heck out of it, effectively disabling thermal regulation.
In the Mean Well powersupply (from what i have found), there is a Class Y capacitor going from 24VDC negative to mains earth, another one going from HVDC (rectified mains AC) to mains earth and third one from 24VDC negative to HVDC negative. The third one will cause slight AC leakage when running from 240VAC. I need to remove that one to isolate the voltage sides from each other. However, this might introduce some unwanted noise, so i'm asking if anyone has done that and what should i do to isolate the sections.
My second option is to build a DIY linear power supply using an old HIFI amp transformer and some salvage parts i have lying around, but that is quite difficult and i don't have experience with building circuits from ground up.

 

[(*steve*)]

if i take my laptop charger and set my MM to 20VAC, i can measure 9.78VAC between positive of the supply and mains earth.

I think you'll find that very little current can follow via this path.

Connect a 10k resistor between the negative output and ground, then measure the voltage across it.

I wasn't aware that these soldering iron had an Earth connection. Is there too electrically connected to the negative supply rail? If so, the resistor I mentioned should be sufficient to keep the top voltage almost at Earth potential.

 

[DaveK]

The TS100 does have a ground connection like this (source: reddit)

This should be the schematic part:

I'm mostly concerned about damaging vintage BJTs that are low noise, low voltage parts. I've even seen leakage so bad it turned on some LEDs while soldering.
People have been using 1M resistors, but i don't have any of those, not even the 10k, i have a 200K one though, can i use that? for both the measurement and then the grounding connection.
Thanks for the help!

 

[(*steve*)]

Yeah, give the 200k a go.

 

[DaveK]

Yeah, give the 200k a go.

So i connected the resistor between mains earth and negative (barrel jacket). The AC component on the positive line shot up to 40VAC when measuring from the jack pin to the same earth spot as the grounding from the negative line.

 

[(*steve*)]

But what happened to the voltage at the negative rail?

And what do you get when you read the output voltage of the power supply on an AC range?

Some cheaper meters (and some not so cheap meters) don't ignore the DC component of a measured voltage on the AC ranges.

 

[DaveK]

But what happened to the voltage at the negative rail?

And what do you get when you read the output voltage of the power supply on an AC range?

Some cheaper meters (and some not so cheap meters) don't ignore the DC component of a measured voltage on the AC ranges.

When i first measured it, i made a mistake and i set the MM to 20VDC, probe the positive rail and other end to mains earth, the MM showed 9.78V. When i did the same with AC range, it showed 13V. Could it mean the AC is only around 3.2V?
Measuring the AC output of the PSU the MM shows 40.3V.

 

[dave9]

Are you going to be using this budget grade soldering iron constantly enough that efficiency makes a difference? Most low cost stations use an old school line isolated transformer not a switcher then earth ground the shaft/tip. If it has its own temperature regulation then you probably don't need a precise, regulated output.

You did mention DIY building something similar but this is not hard to do, and if you look around enough you can probably find something ready made that's not terribly expensive. The one catch is probably that 3A is less common than 2A. 2A, you could just get one of those Hakko clone soldering stations for $20 delivered and throw a bridge rectifier and capacitor in it, bypassing the control board (just an example, get the 220VAC input version if applicable):


https://www.ebay.com/itm/110V-INSTR...C-SOLDERING-STATION-936-US-STOCK/371674334437

 

[(*steve*)]

Could it mean the AC is only around 3.2V?

Yeah, that seems reasonable.

Let's go back to what you said earlier...

i can't have AC leaking from the tip when working on sensitive semiconductors. I obviously have to ground the tip, but the heater is in series with thermocouple and if the leakage from the power supply is too high, it's going to cause high current through the thermocouple and confuse the heck out of it, effectively disabling thermal regulation.

Grounding the tip is fine, but it doesn't actually need to be grounded. It just has to be at the same potential as what you're working on. You could use one rail of your low voltage power supply, as long as your work area is also at this potential. Having said that, mains earth is often used because it *should* always be safe. (Recently a child in my state received a near fatal shock from something connected to mains earth, so it's not a guarantee)

I've not heard of this problem with the thermocouple and mains leakage. What is your evidence that it's actually an issue? Can you provide a link to something?

 

[(*steve*)]

Ok, after the briefest research... I see a long discussion on eevblog and what appears to be the answer.

1. Earth the power supply
2. Earth the iron

Apparently many cheap switchmode power supplies aren't earthed. Correctly earthing them reduces the voltage a lot, and then earthing the tip itself drops the voltage to almost nothing (and the regulation still works).

Edit: after post 12 in that thread they start taking about how nasty some cheap switch mode power supplies are. Posts 13 and 14 also refer to problems with the tips which can either cause it exacerbate the problem. It may be worth a read.

Edit 2: apparently the upgraded "Ralim" firmware also addresses the temperature fluctuation caused by AC leakage.

 

[DaveK]

Ok, after the briefest research... I see a long discussion on eevblog and what appears to be the answer.

1. Earth the power supply
2. Earth the iron

Apparently many cheap switchmode power supplies aren't earthed. Correctly earthing them reduces the voltage a lot, and then earthing the tip itself drops the voltage to almost nothing (and the regulation still works).

Edit: after post 12 in that thread they start taking about how nasty some cheap switch mode power supplies are. Posts 13 and 14 also refer to problems with the tips which can either cause it exacerbate the problem. It may be worth a read.

Edit 2: apparently the upgraded "Ralim" firmware also addresses the temperature fluctuation caused by AC leakage.

Yeah, that is the discussion i am basing off. I've also seen a video of worst-case scenario on youtube where someone touches LED board with the iron and the LEDs light up.
I'm planning to flash the Ralim, so that's good to know, thanks.
For the grounding, my mains earth is safe, i have never received a shock or anything.
The station comes with a power supply, but that's some fishy chinese one, so i better stay away from that.
With the power brick i already have, is it safe to use multimeter continuity and measure if the negative output is already grounded? It has a 3 prong plug and looks to be atleast partially internally grounded.

Are you going to be using this budget grade soldering iron constantly enough that efficiency makes a difference? Most low cost stations use an old school line isolated transformer not a switcher then earth ground the shaft/tip. If it has its own temperature regulation then you probably don't need a precise, regulated output.

You did mention DIY building something similar but this is not hard to do, and if you look around enough you can probably find something ready made that's not terribly expensive. The one catch is probably that 3A is less common than 2A. 2A, you could just get one of those Hakko clone soldering stations for $20 delivered and throw a bridge rectifier and capacitor in it, bypassing the control board (just an example, get the 220VAC input version if applicable):


https://www.ebay.com/itm/110V-INSTR...C-SOLDERING-STATION-936-US-STOCK/371674334437

I already have a transformer (from a hifi amp with 250w power consumption), complete with fuse and mains filtering circuit (i guess it's a filtering circuit). I also have 4 panasonic rectification diodes 200V 3A and bunch of other usable components. My plan was to throw everything on a protoboard with LM338 regulating the voltage. According to a schematic something like this. I don't have experience with building circuits from ground up though, so i might need some help with a wiring diagram.

 

[DaveK]

I think the problem is my multimeter. It is not True RMS, so it reads the mixed ac and dc incorrectly. I guess the delta powersupply leaks very low.