Author Topic: TS100 doesn't function properly when earthed  (Read 7708 times)

cowana, brabus and 2 Guests are viewing this topic.

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
TS100 doesn't function properly when earthed
« on: April 13, 2018, 11:54:05 pm »
Hi all,

I just received my TS100. A known issue with these irons (as discussed on this very forum) is the non-ESD safety / AC voltage on the tip. Most SMPS are not earth-referenced and floating and they tend to leak some AC current. I have a 24V brick and a bench PSU available for powering my TS100. Both have a three prong IEC connectors, are floating and leak AC. Bench PSU ~30VAC to ground and brick is ~80VAC. This propagates to the tip of the TS100. Clearly this needs to be addressed before working on sensitive parts. Thankfully, the TS100 has dedicated earth screw on the body. When I build myself a cable for it I made sure to get a 3-core one so I could attach a cable shoe to the earth terminal.

Now, to my great confusion, the otherwise functional TS100 does not work anymore once a connection between the ground screw and mains earth has been established. The measured temperature just jumps randomly all over the place. Proper temperature readouts resume as soon as the earth connection is removed.

I'd be interested in why this happens, if other people have the same issue and what I can do to get an ESD-safe TS100 setup.

Thanks!
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #1 on: April 14, 2018, 02:57:32 am »
Hi, welcome to the forum.

It sounds as if you have something screwy going on there. Mains adapters with 3-pin IEC connectors almost always have their outputs earthed. Those that don't have their interference suppression component leakage currents returned to mains earth (unless they are cheap fakes of course. It is also very unusual to find a bench PSU that puts out any significant leakage current to earth.

It seems quite likely that the TS100 is being upset by a high supply leakage current when you earth it (I suppose it might be a faulty one but it seems unlikely), the question is, where is it coming from?

The first thing to do is to turn everything off and use a meter to check continuity between the barrel of the mains adapter output connector and the earth pin on its IEC input connector. Do the same with the chassis of the Bench PSU and its IEC connector earth pin. The next thing to check is the earth continuity of the IEC cable(s) themselves.

What were you using as the earth reference when you measured those floating voltages on the Adapter and Bench PSU?  It's possible that this reference is not earthed or has some leakage current on it - there should at least be continuity between this point and the chassis of the Bench PSU.

Hopefully these tests will turn something up.
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
Re: TS100 doesn't function properly when earthed
« Reply #2 on: April 14, 2018, 04:55:11 am »
Thanks!

I bought a 24V/6A brick PSU from Amazon for the TS100. I made sure it's a 3-prong to hopefully get a better unit. It's not like Amazon product reviews tell you about earthing & ripple & y-caps :)

Here:

https://images-na.ssl-images-amazon.com/images/I/61mWltnhadL._SL1000_.jpg

This brick PSU is certainly not mains earth referenced and leaks plenty of AC like most SMPS units. You're right, you might expect something better from a 3-prong device, but I didn't think much about it. Earth is certainly connected to the unit, but since it has no metal case and is one of those plastic welded shut things I can't probe much further.

My bench PSU is an Eventek KPS305D. It's rather low-end, but served me well so far. It's not mains earth referenced either. Has no ground terminal on the front. The output is floating. The case is painted but if I poke around a bit I can find some exposed metal parts that have a low resistance path to mains ground. It also leaks plenty VAC.

My reference earth is mains earth. Earthing on my bench power strip is working as intended. I live in an old house with 2-wire electricity, so we have neutral & earth bridged at the power socket. I know that's bad, I know why that's bad, don't like it, can't change it :)

I must have like a hundred devices here with cheap SMPS that each have ~50% mains VAC between their ground and mains earth. My Anker USB charger, my TV, etc. My point is, I'm surprised that the TS100 is so upset by this. I thought that was the point of an earthing point on the iron. I bet that every single of those SMPSs included in TS100 kits leaks like half mains VAC. Funny thing, if I simply touch the earth pin on a power socket with one hand and and the tip of the TS100 with the other it's enough to make the temperature sensing go crazy.
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #3 on: April 14, 2018, 05:45:02 am »
It sounds like you have a good handle on exactly what is going on then. Regarding the mains supply ground, I can well understand that you're stuck with it. Hopefully you don't have issues with non-bonded plumbing etc.

Personally, I would treat the non-earthed output mains adapter as grounds for return to Amazon, if the output is floating and leaking like that then it suggests that the IEC earth pin isn't actually internally connected. A dodgy sign in my book, who knows what other liberties have been taken inside.  :-\  You would almost certainly be better off with a second-hand major brand name brick off ebay.

For the bench supply, in your general situation, you might want to consider providing a ground lead from a chassis screw so that you can earth the output when you wish/need to do so.

Regarding the TS100 itself. The connector barrel outer (negative) contact is AC decoupled to the tip via a 100nF capacitor (C26, schematic attached). I would have thought 
that would have shunted quite a lot of the (typically 2200pf) Y cap leakage current. I don't know what the voltage rating of C26 is, it's an smd part. There is bound to be some parasitic capacitance between the tip metalwork and the thermocouple / element and a little leakage current too. That might be enough to unsettle the thermocouple sensing, but with C26 shunting the majority of the leakage current, I'm surprised.

Just a though, try checking for DC continuity between the connector barrel and the tip - there's always the possibility of a faulty tip I suppose.

In my iron (with and earthed brick) I have bridged across C26 to provide tip earthing without the inconvenience of a separate earth wire. There is a small voltage on the tip as a result of cable resistance, but it is very low. Obviously not something that you can consider until your PSU leakage issues and stability are under control.

One other suggestion that might work is to change the firmware to Railm's. It uses different (uses more CPU H/W features than the stock f/w) the difference in the control loop might be more resilient, I don't have great hopes of that fixing the problem though. [EDIT: https://github.com/Ralim/ts100 ]

You can probably tell that I am running out of ideas at this point...
« Last Edit: April 14, 2018, 05:49:22 am by Gyro »
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline sn4k3

  • Frequent Contributor
  • **
  • Posts: 298
  • Country: pt
Re: TS100 doesn't function properly when earthed
« Reply #4 on: April 14, 2018, 06:49:58 am »
That cheap power supply sucks, and often not earthed, even when they have 3 pins on board, most of them just connect two wires, so the 3 pin cable is just to lie to the customer.
I recommend a DIY PSU: https://www.thingiverse.com/thing:2847365 that way you know what you have.
You can also try to direct connect the TS100 earth to earth socket and see if same wierd things happen. Or open your power supply and check connections and build
 

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
Re: TS100 doesn't function properly when earthed
« Reply #5 on: April 14, 2018, 07:04:09 am »
I like your 2-wire solution. I didn't want to order some special soldering iron cord, so I just got the thinnest silicone 3-core cable I could find that still had conductors with appropriate diameter for the current. It's a bit heavy, would be much easier to find a good heat resistant & flexible 2-core cable

Had a quick look through the house, I don't think I own any other earthed PSU. In my entertainment setup everything is 2-prong SMPS & leaking AC to the point where I grounded my video switcher as I was tired of getting zapped :o

So, how is a properly grounded PSU actually wired? Simply a bridge between mains earth and DC out negative? Can't I just do this myself at the output? I can return mine to Amazon, but I'm not even sure how/where to buy such a power supply, they seem highly exotic. Can't find anything good looking locally on eBay and I think the one I bought was the only one Amazon sold with a three prong connector. So it's a bit strange that the TS100 would expect a grounded PSU (which 99% of 24V PSUs seemingly aren't), especially if it has an earthing point that you apparently can't actually use until you got one of those ;D

I can actually measure 100nF between negative input & tip, all good!

sn4k3 - What do you mean by directly connect the TS100 earth to earth socket? Unless I misunderstand you that's exactly what I've been doing. I connect the earthing screw on the TS100 to mains earth. Even just touch the earth screw with my finger causes the temperature to start fluctuating since my body is obviously reasonably close to mains earth.

Edit: I just tried tying DC out negative on my PSU to mains earth, still got ~10VAC potential to mains earth on the iron tip. But now if I earth the tip through the ground screw, temperature sensing continues to work fine. Interesting.
« Last Edit: April 14, 2018, 07:21:42 am by Rooster Cogburn »
 

Offline TheAmmoniacal

  • Supporter
  • ****
  • Posts: 1120
  • Country: no
  • Cut me some slack, I'm a biochemist!
Re: TS100 doesn't function properly when earthed
« Reply #6 on: April 14, 2018, 07:40:59 am »
Be careful running the TS100 with a 6A capable supply, even at 24V, you're likely to blow out the MOSFET quickly.
I collect [corporate] mugs.
MTBF ~ 700.000 h
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #7 on: April 14, 2018, 07:43:08 am »
It's a bit heavy, would be much easier to find a good heat resistant & flexible 2-core cable

Yes, you can find it on ebay (where else!  :D)

Quote
So, how is a properly grounded PSU actually wired? Simply a bridge between mains earth and DC out negative? Can't I just do this myself at the output? I can return mine to Amazon, but I'm not even sure how/where to buy such a power supply, they seem highly exotic. Can't find anything good looking locally on eBay and I think the one I bought was the only one Amazon sold with a three prong connector. So it's a bit strange that the TS100 would expect a grounded PSU (which 99% of 24V PSUs seemingly aren't), especially if it has an earthing point that you apparently can't actually use until you got one of those ;D

Yes, exactly that, just mains ground to output negative. As I say, I'm suspicious of 3-pin bricks that don't do this. As far as the TS100 requiring a grounded PSU, this isn't an issue that I've seen being flagged, quite a few reports of poor quality bricks fading and causing a reset when the heater powers up (not your case - those go back to the start screen) but I don't remember yours being a common problem.

Quote
Edit: I just tried tying DC - on my PSU to mains earth, still got ~10VAC potential to mains earth on the tip. But now if I earth the tip, temperature sensing works fine. Interesting.

That's curious, unless your meter is measuring some 'phantom' effect then I still get the suspicion that there's something screwy going on with your mains earth connection (aside from your incoming 2-wire supply issue). There's a very good connection between the earthing screw and tip (again, check continuity) so the 10VAC can't be coming from there. It sounds more like the voltage drop across a length of conductor carrying significant current, (maybe a high resistance joint in the mains neutral/earth wiring?). Best to proceed with caution, particularly if tempted to use the current range of your meter!
« Last Edit: April 14, 2018, 07:48:20 am by Gyro »
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #8 on: April 14, 2018, 07:45:23 am »
Be careful running the TS100 with a 6A capable supply, even at 24V, you're likely to blow out the MOSFET quickly.

I believe that problem has only been reported once - with a faulty element. I've been running mine at 24V for a long time now.
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline TheAmmoniacal

  • Supporter
  • ****
  • Posts: 1120
  • Country: no
  • Cut me some slack, I'm a biochemist!
Re: TS100 doesn't function properly when earthed
« Reply #9 on: April 14, 2018, 07:47:51 am »
Be careful running the TS100 with a 6A capable supply, even at 24V, you're likely to blow out the MOSFET quickly.

I believe that problem has only been reported once - with a faulty element. I've been running mine at 24V for a long time now.

Mine will go above the rated output power if I don't current-limit the input, it's possible newer revisions have fixed that; but this video is from March
I collect [corporate] mugs.
MTBF ~ 700.000 h
 

Offline sn4k3

  • Frequent Contributor
  • **
  • Posts: 298
  • Country: pt
Re: TS100 doesn't function properly when earthed
« Reply #10 on: April 14, 2018, 07:53:24 am »
I like your 2-wire solution. I didn't want to order some special soldering iron cord, so I just got the thinnest silicone 3-core cable I could find that still had conductors with appropriate diameter for the current. It's a bit heavy, would be much easier to find a good heat resistant & flexible 2-core cable

Had a quick look through the house, I don't think I own any other earthed PSU. In my entertainment setup everything is 2-prong SMPS & leaking AC to the point where I grounded my video switcher as I was tired of getting zapped :o

So, how is a properly grounded PSU actually wired? Simply a bridge between mains earth and DC out negative? Can't I just do this myself at the output? I can return mine to Amazon, but I'm not even sure how/where to buy such a power supply, they seem highly exotic. Can't find anything good looking locally on eBay and I think the one I bought was the only one Amazon sold with a three prong connector. So it's a bit strange that the TS100 would expect a grounded PSU (which 99% of 24V PSUs seemingly aren't), especially if it has an earthing point that you apparently can't actually use until you got one of those ;D

I can actually measure 100nF between negative input & tip, all good!

sn4k3 - What do you mean by directly connect the TS100 earth to earth socket? Unless I misunderstand you that's exactly what I've been doing. I connect the earthing screw on the TS100 to mains earth. Even just touch the earth screw with my finger causes the temperature to start fluctuating since my body is obviously reasonably close to mains earth.

Edit: I just tried tying DC out negative on my PSU to mains earth, still got ~10VAC potential to mains earth on the iron tip. But now if I earth the tip through the ground screw, temperature sensing continues to work fine. Interesting.

You can try mains earth and DC out negative, some designs use that and recommend it as well when proper earth is missing.
Also try a 1MOhm resistor in series from earth to TS100 earth
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #11 on: April 14, 2018, 07:58:04 am »
Be careful running the TS100 with a 6A capable supply, even at 24V, you're likely to blow out the MOSFET quickly.

I believe that problem has only been reported once - with a faulty element. I've been running mine at 24V for a long time now.

Mine will go above the rated output power if I don't current-limit the input, it's possible newer revisions have fixed that; but this video is from March

Interesting - no, mine is a first version one, from last year sometime. I use the Ralim firmware, maybe that controls the heater current better (the MOSFET drive is AC coupled anyway, so it's impossible to turn it on continuously whichever firmware). It might also be a new version problem, MOSFET part change? Worth looking out for in that case.
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
Re: TS100 doesn't function properly when earthed
« Reply #12 on: April 14, 2018, 09:29:16 am »
Yes, exactly that, just mains ground to output negative. As I say, I'm suspicious of 3-pin bricks that don't do this. As far as the TS100 requiring a grounded PSU, this isn't an issue that I've seen being flagged, quite a few reports of poor quality bricks fading and causing a reset when the heater powers up (not your case - those go back to the start screen) but I don't remember yours being a common problem.

I actually think mine has reset a couple of times. I just got the iron, so I'm not used to what it does when it goes into sleep mode etc. Looks like the PSU I bought is a total PoS, I'll keep going with my bench supply till I got a better one. I also haven't updated the firmware yet (I'll likely end up running the open source replacement one you linked).

Quote
Edit: I just tried tying DC - on my PSU to mains earth, still got ~10VAC potential to mains earth on the tip. But now if I earth the tip, temperature sensing works fine. Interesting.

That's curious, unless your meter is measuring some 'phantom' effect then I still get the suspicion that there's something screwy going on with your mains earth connection (aside from your incoming 2-wire supply issue). There's a very good connection between the earthing screw and tip (again, check continuity) so the 10VAC can't be coming from there. It sounds more like the voltage drop across a length of conductor carrying significant current, (maybe a high resistance joint in the mains neutral/earth wiring?). Best to proceed with caution, particularly if tempted to use the current range of your meter!

I just bought a replacement fuse 400mA ;-) I checked the wiring, the neutral/earth bridge is working correctly. I measured again and to be more clear:

- no grounding on iron & no grounding PSU: ~80VAC to mains earth
- no grounding on iron & grounding PSU: ~1.5VAC to mains earth
- grounding on iron & grounding PSU: ~0.015 VAC to mains earth

So that looks good? Actually, isn't option 3 just what you did? You got a grounded PSU and bridged C26 to get negative on the tip, I'm just doing it with three wires and attach to the tip through the ground screw.

You can try mains earth and DC out negative, some designs use that and recommend it as well when proper earth is missing.
Also try a 1MOhm resistor in series from earth to TS100 earth

This also won't work. If I just touch the earth screw on the iron the temperature measurements start fluctuating. Gets worse when I ground myself with mains earth. So it seems that if you have a PSU with significant AC leakage you can't just ground the tip without screwing with the thermocouple or however that works (I admit I have no clue about how soldering iron heating elements actually work...).

Is there a quality, ready-build, earthed brick PSU I can buy somewhere in Europeland? I don't want to build a kit as SMPSs are quite complicated and I don't trust myself with mains voltage yet :)
 

Offline MacMeter

  • Frequent Contributor
  • **
  • Posts: 316
  • Country: us
Re: TS100 doesn't function properly when earthed
« Reply #13 on: April 14, 2018, 12:41:15 pm »
Be careful running the TS100 with a 6A capable supply, even at 24V, you're likely to blow out the MOSFET quickly.

I believe that problem has only been reported once - with a faulty element. I've been running mine at 24V for a long time now.

Mine will go above the rated output power if I don't current-limit the input, it's possible newer revisions have fixed that; but this video is from March

I had a brief back and forth with that video author, after seeing the same video in another thread here:
https://www.eevblog.com/forum/beginners/choosingunderstanding-the-right-powersuply-for-the-ts100/msg1473804/#msg1473804
 

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
Re: TS100 doesn't function properly when earthed
« Reply #14 on: April 14, 2018, 07:14:47 pm »
I had a brief back and forth with that video author, after seeing the same video in another thread here:
https://www.eevblog.com/forum/beginners/choosingunderstanding-the-right-powersuply-for-the-ts100/msg1473804/#msg1473804

I think I saw exactly that and made sure to test my tips before powering on, seems all good so far!
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #15 on: April 14, 2018, 07:33:57 pm »
I had a brief back and forth with that video author, after seeing the same video in another thread here:
https://www.eevblog.com/forum/beginners/choosingunderstanding-the-right-powersuply-for-the-ts100/msg1473804/#msg1473804

Good to know, thanks. I haven't experienced a bad tip yet. A shorted tip would certainly be a good reason for a shorted mosfet with a high current supply!
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #16 on: April 14, 2018, 08:37:26 pm »
I just bought a replacement fuse 400mA ;-) I checked the wiring, the neutral/earth bridge is working correctly. I measured again and to be more clear:

- no grounding on iron & no grounding PSU: ~80VAC to mains earth
- no grounding on iron & grounding PSU: ~1.5VAC to mains earth
- grounding on iron & grounding PSU: ~0.015 VAC to mains earth

So that looks good? Actually, isn't option 3 just what you did? You got a grounded PSU and bridged C26 to get negative on the tip, I'm just doing it with three wires and attach to the tip through the ground screw.

Yes, that all looks good. Grounding the tip by bridging C26 is almost as good as your option 3, the difference is that the tip 'sees' the voltage drop across the negative conductor (due to heater current). on my setup it results in ~0.1 to 0.2V which is low enough for me to be comfortable. Using 3 cores and the ground screw is the ultimate - but less convenient.

Quote
Is there a quality, ready-build, earthed brick PSU I can buy somewhere in Europeland? I don't want to build a kit as SMPSs are quite complicated and I don't trust myself with mains voltage yet :)

There are some Chinese brands that are much higher quality than others (effectively main brand) these include Meanwell, Delta, LTE etc. Avoid anything that doesn't have a brand name. As I mentioned previously, a second hand main brand Sony, Dell, Lenovo, HP etc. off ebay would be good. You might have to sacrifice a couple of volts but I think some are 23V 90W.
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 2526
  • Country: fr
Re: TS100 doesn't function properly when earthed
« Reply #17 on: April 15, 2018, 12:18:54 am »
That cheap power supply sucks, and often not earthed, even when they have 3 pins on board, most of them just connect two wires, so the 3 pin cable is just to lie to the customer.
I recommend a DIY PSU: https://www.thingiverse.com/thing:2847365 that way you know what you have.
You can also try to direct connect the TS100 earth to earth socket and see if same wierd things happen. Or open your power supply and check connections and build

So you recommend him to 3D print an enclosure for a random no-name $7 SMPS module from China?  :palm:

The build quality of those modules is plainly visible from the photos already - even if those Nichicon caps are genuine (unlikely, given the price of the entire thing), they are neatly tucked inside of the heatsink so they will last long for sure like that. The connectors are mounted crooked, etc. The second Geekcreit branded one is a bit better but only slightly.

So yes, with that the OP will certainly know what they are getting. OP can suspect that his supply from Amazon is crap however if they follow your recommendation, they will know it being one for sure.

Totally ridiculous considering you can get a decent, known safe, supply for 50 bucks or so:
http://uk.farnell.com/xp-power/veh150ps24/power-supply-150w-24v-6-25a/dp/2365130

How much is your safety and your house/apartment worth that you would rely on a $7 mains module like this?

« Last Edit: April 15, 2018, 12:25:17 am by janoc »
 

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
Re: TS100 doesn't function properly when earthed
« Reply #18 on: April 15, 2018, 12:27:27 am »
Quote
Is there a quality, ready-build, earthed brick PSU I can buy somewhere in Europeland? I don't want to build a kit as SMPSs are quite complicated and I don't trust myself with mains voltage yet :)

There are some Chinese brands that are much higher quality than others (effectively main brand) these include Meanwell, Delta, LTE etc. Avoid anything that doesn't have a brand name. As I mentioned previously, a second hand main brand Sony, Dell, Lenovo, HP etc. off ebay would be good. You might have to sacrifice a couple of volts but I think some are 23V 90W.

Thanks, great advice. Had a look inside my power brick, looks bad, earth is not even connected:

https://i.imgur.com/qMdxawf.png

 ::)

The Mean Well PSUs seem readily available at decent prices. I found a spec sheet for the 90W ones:

https://media.it-tronics.de/Datasheets/Power_Supplies/MeanWell/GST90A.pdf

It says 'Class I power (with earth pin)' and 'LEAKAGE CURRENT (max.) 1mA/240VAC'. That seems like a lot. Can I somehow tell from the spec sheet that the output is actually mains earth referenced?
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 2526
  • Country: fr
Re: TS100 doesn't function properly when earthed
« Reply #19 on: April 15, 2018, 12:34:04 am »
Is there a quality, ready-build, earthed brick PSU I can buy somewhere in Europeland? I don't want to build a kit as SMPSs are quite complicated and I don't trust myself with mains voltage yet :)

One option are laptop power bricks - however those output around 19-20V usually. Can be cheaply had from eBay and a lot of other places.

If you want new, just look at RadioSpares/RS, Farnell, Mouser and similar places. All are selling these kinds of power bricks in a range of output voltages and powers. Just choose one with the 3 pin IEC mains cable connector to have the earth connection (e.g. the XP power one I have linked in the post above).

Quote
It says 'Class I power (with earth pin)' and 'LEAKAGE CURRENT (max.) 1mA/240VAC'. That seems like a lot. Can I somehow tell from the spec sheet that the output is actually mains earth referenced?

My guess is that it isn't - page 3, description of the output DC jack says:
"-V not connected to AC FG".



« Last Edit: April 15, 2018, 12:38:08 am by janoc »
 

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
Re: TS100 doesn't function properly when earthed
« Reply #20 on: April 15, 2018, 12:49:21 am »
The problem is that just because it has a 3-prong input does not mean it has ground tied to mains earth or that the pin is even connected, see the picture of the brick I bought from Amazon or the Mean Well PSU. Seems like this is a rather rare design choice and hard to find.

The spec sheet of the supply you linked is neat:

https://www.prosoft.ru/cms/f/458667.pdf

Actually says 'Negative output is connected to Ground'.

After shipping it costs about as much as two TS100s, though ;-)
« Last Edit: April 15, 2018, 01:05:46 am by Rooster Cogburn »
 

Online Gyro

  • Super Contributor
  • ***
  • Posts: 3668
  • Country: gb
Re: TS100 doesn't function properly when earthed
« Reply #21 on: April 15, 2018, 02:05:52 am »
Quote
Is there a quality, ready-build, earthed brick PSU I can buy somewhere in Europeland? I don't want to build a kit as SMPSs are quite complicated and I don't trust myself with mains voltage yet :)

There are some Chinese brands that are much higher quality than others (effectively main brand) these include Meanwell, Delta, LTE etc. Avoid anything that doesn't have a brand name. As I mentioned previously, a second hand main brand Sony, Dell, Lenovo, HP etc. off ebay would be good. You might have to sacrifice a couple of volts but I think some are 23V 90W.


Thanks, great advice. Had a look inside my power brick, looks bad, earth is not even connected:

https://i.imgur.com/qMdxawf.png

 ::)

The Mean Well PSUs seem readily available at decent prices. I found a spec sheet for the 90W ones:

https://media.it-tronics.de/Datasheets/Power_Supplies/MeanWell/GST90A.pdf

It says 'Class I power (with earth pin)' and 'LEAKAGE CURRENT (max.) 1mA/240VAC'. That seems like a lot. Can I somehow tell from the spec sheet that the output is actually mains earth referenced?

Ugh, that brick is nasty inside, I think they forgot that they needed to include space for the input side reservoir capacitor!  :palm: Hard to tell what's lurking underneath it.  Top side it's a mess - the led and primary-secondary Y-cap are pressed up against the primary side heatsink. The common mode choke is pressed right up against the AC input connector - no wonder the earth pin isn't connected. At first glance they seem to have had a reasonable stab at creepage and clearance distances on the bottom side... That's until you notice that the secondary side heatsink mounting tab is soldered within a few mm of the primary side tracks, nullifying all of the other clearances!  :scared: No attempt at reinforced insulation on the heatsinks where they come close to components. Without an earth connection it should meet the requirements of a Class II double insulated product - sadly it misses by a mile! Annotated photo attached for interest/education.

I'd send it back as very unsafe! Maybe report it to whatever your equivalent of Trading Standards is too.  An instructive teardown though, thanks!

It's a shame the Meanwell has an isolated output (the diagram does seem to indicate that). Regarding the 1mA leakage current, that is listed under the 'Input' section, I would hope that it means maximum leakage to the mains earth pin, not to the output. It may be possible to check with them, as I indicated, they are a reputable manufacturer.

For now, you might just want to pick up a second-hand 20V 65W/90W Lenovo brick from ebay. True, it will drop the maximum wattage, but the tip is well regulated so you probably won't notice in practice. The heater rarely runs flat-out anyway. Best of all, they can be had for a few Euro on ebay (make sure you pick a 3 pin input version).


EDIT: The annoying thing is that, with a little extra effort, some additional insulation and a securely fixed connection between the Input Earth pin and the Output Ground, they could have had a reasonably safe (though still nasty) product. Fundamental lack of understanding or caring I guess!  ::)

EDIT1: Image updated. P=Primary side heatsink, S=Secondary side.
« Last Edit: April 15, 2018, 05:27:28 am by Gyro »
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline sn4k3

  • Frequent Contributor
  • **
  • Posts: 298
  • Country: pt
Re: TS100 doesn't function properly when earthed
« Reply #22 on: April 15, 2018, 04:59:51 am »
That cheap power supply sucks, and often not earthed, even when they have 3 pins on board, most of them just connect two wires, so the 3 pin cable is just to lie to the customer.
I recommend a DIY PSU: https://www.thingiverse.com/thing:2847365 that way you know what you have.
You can also try to direct connect the TS100 earth to earth socket and see if same wierd things happen. Or open your power supply and check connections and build

So you recommend him to 3D print an enclosure for a random no-name $7 SMPS module from China?  :palm:

The build quality of those modules is plainly visible from the photos already - even if those Nichicon caps are genuine (unlikely, given the price of the entire thing), they are neatly tucked inside of the heatsink so they will last long for sure like that. The connectors are mounted crooked, etc. The second Geekcreit branded one is a bit better but only slightly.

So yes, with that the OP will certainly know what they are getting. OP can suspect that his supply from Amazon is crap however if they follow your recommendation, they will know it being one for sure.

Totally ridiculous considering you can get a decent, known safe, supply for 50 bucks or so:
http://uk.farnell.com/xp-power/veh150ps24/power-supply-150w-24v-6-25a/dp/2365130

How much is your safety and your house/apartment worth that you would rely on a $7 mains module like this?

I think you dont have understood my suggestion. The case yes and wiring for better handle mobility and turn TS100 into a fixed station [Looks, usability and convenience]. The psu however you can use your own and is self picked, The recommended PSU on that post is for people who need something that work and cheap, but they can choose whatever they fell best (If fit inside).
I have built some solder station using that power supplies and very people does the same with T12, and so far i can't say bad about them, they work, nothing bad happen so far, reliable and is within spec, the feedbacks are very positive with the Geekcreit. About security i have tested them too, i have blow 4 in total (2 modules each), i have shorted mains 220v in a test porpuse and nothing bad happen (No fire, no burn, no flying pieces, no external damage outside the module) the fuse have blow, two burn resistors and the transistors shorted on the 9A version, again no fire, no burn pcb or serious damage but psu become unusable of course and not worth the repair, still i consider this kinda of safe for the use, i have seen laptop original branded DC bricks cath fire in front of me only by normal use so i think nothing is 100% safe in the real world. At DC side shorting does nothing, after remove the short the system starts normally and the current metter or iron are not affected.
A solder station is a kind of equipament that you won't let it ON while not using for a while or not present for long, so the risk is even smaller, the case have a Mains switch that is very pratical.
If you are still not convinced with any china module the case allow you to connect a external DC power supply to the DC jack, so i can't see nothing wrong with this setup. As bonus you can fit that trusty DC brick inside the case if you want to hide it.
« Last Edit: April 15, 2018, 05:18:16 am by sn4k3 »
 

Offline MacMeter

  • Frequent Contributor
  • **
  • Posts: 316
  • Country: us
Re: TS100 doesn't function properly when earthed
« Reply #23 on: April 15, 2018, 05:56:55 am »
Quote
Is there a quality, ready-build, earthed brick PSU I can buy somewhere in Europeland? I don't want to build a kit as SMPSs are quite complicated and I don't trust myself with mains voltage yet :)

There are some Chinese brands that are much higher quality than others (effectively main brand) these include Meanwell, Delta, LTE etc. Avoid anything that doesn't have a brand name. As I mentioned previously, a second hand main brand Sony, Dell, Lenovo, HP etc. off ebay would be good. You might have to sacrifice a couple of volts but I think some are 23V 90W.

Thanks, great advice. Had a look inside my power brick, looks bad, earth is not even connected:

https://i.imgur.com/qMdxawf.png

 ::)

The Mean Well PSUs seem readily available at decent prices. I found a spec sheet for the 90W ones:

https://media.it-tronics.de/Datasheets/Power_Supplies/MeanWell/GST90A.pdf

It says 'Class I power (with earth pin)' and 'LEAKAGE CURRENT (max.) 1mA/240VAC'. That seems like a lot. Can I somehow tell from the spec sheet that the output is actually mains earth referenced?

That’s the one I’m using:
Mean Well GST90A24-P1M Regulated Switching Table Top Power Supply, 24 Volt, 3.75 Amp, 90 Watt.

NOTE: Did not come with power cord, I had a US 3-wire grounded cord.
 

Offline Rooster Cogburn

  • Regular Contributor
  • *
  • Posts: 52
  • Country: de
Re: TS100 doesn't function properly when earthed
« Reply #24 on: April 15, 2018, 08:57:39 am »
Ugh, that brick is nasty inside, I think they forgot that they needed to include space for the input side reservoir capacitor!  :palm: Hard to tell what's lurking underneath it.  Top side it's a mess - the led and primary-secondary Y-cap are pressed up against the primary side heatsink. The common mode choke is pressed right up against the AC input connector - no wonder the earth pin isn't connected. At first glance they seem to have had a reasonable stab at creepage and clearance distances on the bottom side... That's until you notice that the secondary side heatsink mounting tab is soldered within a few mm of the primary side tracks, nullifying all of the other clearances!  :scared: No attempt at reinforced insulation on the heatsinks where they come close to components. Without an earth connection it should meet the requirements of a Class II double insulated product - sadly it misses by a mile! Annotated photo attached for interest/education.

I'd send it back as very unsafe! Maybe report it to whatever your equivalent of Trading Standards is too.  An instructive teardown though, thanks!

It's a shame the Meanwell has an isolated output (the diagram does seem to indicate that). Regarding the 1mA leakage current, that is listed under the 'Input' section, I would hope that it means maximum leakage to the mains earth pin, not to the output. It may be possible to check with them, as I indicated, they are a reputable manufacturer.

For now, you might just want to pick up a second-hand 20V 65W/90W Lenovo brick from ebay. True, it will drop the maximum wattage, but the tip is well regulated so you probably won't notice in practice. The heater rarely runs flat-out anyway. Best of all, they can be had for a few Euro on ebay (make sure you pick a 3 pin input version).


EDIT: The annoying thing is that, with a little extra effort, some additional insulation and a securely fixed connection between the Input Earth pin and the Output Ground, they could have had a reasonably safe (though still nasty) product. Fundamental lack of understanding or caring I guess!  ::)

EDIT1: Image updated. P=Primary side heatsink, S=Secondary side.

Yeah, sending that PSU back. I've build a few linear PSUs with transformer, rectifier, filtering cap, 78xx regulator etc., but haven't looked too deeply into SMPSs yet. I now have some examples of what not to do 8)

I've gotten some very good suggestions for PSUs, thanks everyone! Tomorrow is flea market here, I'll see if I can find a 3-prong laptop power brick from Lenovo, Dell etc. If not, I'll either order one from eBay or shell out the cash for one of the Mean Well / XP Power ones.


I did some more experiments with the firmware. I tried the well-known firmware from Ralim, but I'm having this bug: https://github.com/Ralim/ts100/issues/261. Basically, the temperature climbs smoothly to the set one and then randomly fluctuates ~30C. Can't really use it. I switched back to the official firmware and noticed something peculiar. The temperature readout fluctuations I described when I ground the tip without also grounding the PSU only happen once I set a temperature >300C. Anything below is completely stable. Only goes crazy above it. Really weird.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf