PSU issue... need a bit assistance...

Electronics talk! Arduino UNO, Raspberry Pi and more
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

PSU issue... need a bit assistance...

Post by drayson »

Ladies, Gentleman...

I have to bother you with some PSU questions... I honestly completely out of ideas... maybe somebody else has a clue...

So:
I have a self-build heated bed running nice on 12V/10A 120W, but does not go over 65°C... max. 70 after 13min heating...
So this is what I´d like to change as I´d like to print also with other materials with a need for higher bed temps.

Here's the sellers description (unfortunaltey not available any more at ebay):
215 x 215mm Ni-Cr Silicone self-adhesive heater
This heater can be used in different modes.
Low temperature - at 12V 8A 96W about 60 deg C
High temperature - at 18V 12A 216W must be regulated

I asked him if a 19,5V 16,9A PSU would also be good and I got a clear YES so I ordered a DELL 330W PSU (http://www.laptopbattery-fan.co.uk/inde ... s_id=17360) with that specs...

... soldered everything together with the relais setup and...

... as soon as the HB get on, the PSU shuts down.
I crosschecked it also with another relais as well as with a MOSFET setup from jonny, but everytime the same thing happens... shutdown... power off, power on, it's here again, but as soon the HB gets on it... right, powers down...

Any ideas?? I´m out of... Any hint/assistance appreciated...
User avatar
jonnybischof
Reactions:
Posts: 310
Joined: February 15th, 2016, 2:48 am
Location: Zurich, Switzerland
3D Printer(s): UMO

Re: PSU issue... need a bit assistance...

Post by jonnybischof »

That sounds like an overcurrent shutdown inside the PSU (supposing there is no basic wiring problem).

So, let's calculate:

The heatbed is nothing other than a resistor: 12 V / 8 A = 1.5 Ohms
Using P = U*U/R, I get 253.5 W at 19.5V. So a 330W PSU should be okay.

So, either your PSU is faulty (unlikely if it's a new one, but possible), or your heater acts up.

The heater works as intended at 12 V? Even now after you played with 19.5 V?
These silicone heaters can be complete rubbish. Mine was broken more or less from the beginning, having a slack-joint on the cable that made operation very... interesting... while the bed was moving up and down in the printer.

Try putting a similar load on the PSU and see how it behaves. That is easier said than done, though. You could get a bunch of 100W resistors and put them in parallel (you want 1.5 Ohms target resistance), or, maybe the simpler approach, use another heatbed. A 12V PCB heatbed (dual mode is ok as well) should put the same load on the PSU as that silicone heater.

Also, you can hook up the heater directly to the PSU and rule out any wiring or relay issues. Just don't leave it powered up any longer than 2-3 minutes...

Using another heatbed hooked up directly to the PSU should reveal your culprit:
a) the PSU fails -> PSU is bad
b) everything fine -> silicone heater is bad (check silicone heater hooked up directly!)
c) if everything is fine when you hook up your silicone heater directly to the PSU, then your problem lies in the wiring or the relay, or the control signal
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

jonny, you are great... thank you for support.

I re-soldered yesterday evening the 12V PSU and it works as usual. app. 5 min. heatup time to 55°C... as slow as usual.
I don´t think that´s the wiring of the relais as I also tried it with your famous MOSFET driver with the same result.

I will hook it directly within the next days and will report the outcome...
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

Re-thinking the situation makes me think the only place of failure can be the heater itself...
Everything soldered correctly, and I forgot to mention that I tried also this PSU http://www.ebay.at/itm/262536144197?_tr ... EBIDX%3AIT without any luck - but I was thinking it was the fault of the PSU....

ok, in an act of thoughtlessly I ordered new parts...

Silicon heater 12V 16,6A 200W
http://www.ebay.at/itm/282127420961?_tr ... EBIDX%3AIT

PSU 12V 18A
http://www.ebay.at/itm/DELL-D222P-01-Ne ... Sw3mpXF69R

damn - planned to spend the money in addons for my new unimat 3 lathe/mill... (http://www.nielsmachines.com/de/emco-ve ... chine.html) sweet - isnt?? :-)
User avatar
jonnybischof
Reactions:
Posts: 310
Joined: February 15th, 2016, 2:48 am
Location: Zurich, Switzerland
3D Printer(s): UMO

Re: PSU issue... need a bit assistance...

Post by jonnybischof »

You're welcome ;)
That lathe looks nice indeed. Haven't had much luck with lathes myself. I can't recommend the OptiTurn TU1503. It's just too small and brittle to be of much use..

Is there a special reason why you stick to silicone heaters? I would recommend a PCB heater, ideally something like the MK3 alu heatbed which is pretty good.
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

Honestly, the only reason is, that I´m too lazy to change my whole bed setup.
I organized a 5mm alu plate in our company workshop which fits dirctly to the UMO mounting pattern, my glass plates and all the other little gimicks I added and so it´s the simplest thing to just change the pad glued onto it.

Yea, the OptiTurn are nice but I understand you point. I decided on the Unimat 3 as it´s made in the 80th in Austria so no China quality. It is of course small, but very reliable (acc. to the reviews) and fit perfectly onto the desk beside my UMO :-)
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

... ok... back again on this topic...

Guys, jonny, all other... it´s frustrating... bought a new PSU, a new silicon heater... soldered everything... DAMN.. same issue again.
As soon as I plug in the power, the PSU struggles and shut down...

Could it be that there is a safety fuse in this damn DELL PSU's which are missing in low budget china crackers? I wonder if there is a way to hack/bypass this...

Beside this, I´m out of ideas... any hint is appreciated...
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

PS: is there any way to generate something like a soft start? Could this work with PID instead of BangBang mode?
Honestly, I´m so frustrated.... :-(
User avatar
jonnybischof
Reactions:
Posts: 310
Joined: February 15th, 2016, 2:48 am
Location: Zurich, Switzerland
3D Printer(s): UMO

Re: PSU issue... need a bit assistance...

Post by jonnybischof »

Did you buy the same Dell PSU again?

Soft start would be possible, but not "that" easy to do, and usually not necessary... The soft start wouldn't be the best idea though if you're switching the bed on and off often (lots of energy loss and heating where you don't want it).

Like I said, I would try using a "real" power supply instead of one specifically made for powering / charging laptops.

19 V is a bit of an issue though as that's not an industry standard voltage.

I would try the Mean Well LRS-350-15, with 15 V output -> ~150 W on your heatbed

Mouser has only one 18 V (~216 W) power supply that would possibly be interesting: TDK CUS200M18. It's much more expensive because it's a more exotic type, and it has medical certification.

If you go 24 V, you will have ~384 W which is an enormous amount of power. Personally not recommending that.
Mean Well HRP-450-24 (or HRPG-450-24). As you can see - big price difference...

I'll be using that HRPG-450-24 brick to run an entire printer with 300x200mm heated bed and 5 high power printheads (IF it gets done some day...).

As you can see... Power supplies aren't as cheap as Ebay wants to tell you. If you buy a 300W PSU for 30$, then it's probably crap.
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

jonnybischof wrote:Did you buy the same Dell PSU again?
No - it´s another/smaller one - 12V 18A for the new 12V heater (the other one has 19,5V 16A for the "old" heater pad)

I already tried "real" ones, also from Meanwell, and one from Lambda but also with the same result. A colleague here in office mentioned that It seems that the starting resistance is too small so it thinks it gets short cutted...

Currently 4 PSU's are bought, laying around and are more or less useless :-(

I will order an additional one - the one you suggested with 15V.. we will see...
User avatar
jonnybischof
Reactions:
Posts: 310
Joined: February 15th, 2016, 2:48 am
Location: Zurich, Switzerland
3D Printer(s): UMO

Re: PSU issue... need a bit assistance...

Post by jonnybischof »

Hmm... I'm not sure how these silicone heaters are actually constructed. Maybe it's an NTC material that increases it's resistance with temperature. Meaning when it's cold, its resistance is very low and might indeed cause the PSU to go into overcurrent protection mode.
Does the heater have something like a nominal temperature?

In that case you'd have to use an unregulated power supply constructed from a transformer coil and a rectifier. And then make sure to use the very best MosFET for control that you can find (at least 30% higher voltage rating than your nominal voltage, and an RDSon in the single digit milliohms range. (typically 250+ A).
And of course you must absolutely use a glass fuse because an unregulated power supply has no protection whatsoever. So, use a very slow blow fuse rated at 125% or 150% of your nominal current (250 V).

Or - probably - use a regulated (switching) power supply that has a much higher load rating.


/edit:
I just noticed that you're now using a 12 V, 16.7 A heatbed not 12 V, 8 A anymore. That means you should absolutely and only use 12V power supplies! My wattage calculations in the above post are bullshit :P

/edit:
PID mode would theoretically get you a soft start (but Marlin doesn't work that way, it does it the other way round - full current when heating up and then slowing down to regulate). But in practice, what happens with PID is that your MosFET (or relay) will be used as a variable resistor by quickly switching it off and on. That means the MosFET or relay will blow out within seconds, unless they have very good heatsinking and active cooling.

/edit:
Looking at your last ebay link for the silicone heater, I don't think it's an NTC based self-regulating heater. But you never really know with that chinese bulk order stuff...

/edit:
From your first post: "215 x 215mm Ni-Cr Silicone self-adhesive heater" Ni-Cr is in fact an NTC material, but the resistance change is very small (a few % from 20 to 300°C) so I don't really see a good reason why your PSUs are failing...

Can you measure the actual resistance of the heater pad with a multimeter?
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

jonnybischof wrote: Does the heater have something like a nominal temperature?
unfortunately not - or better said, I have not found any data on it...
jonnybischof wrote: /edit:
I just noticed that you're now using a 12 V, 16.7 A heatbed not 12 V, 8 A anymore. That means you should absolutely and only use 12V power supplies! My wattage calculations in the above post are bullshit :P
Yes, I have the intention to changed the heat be/silicon heater. The 12V/8A is still muonted, but the 12V/16,6A is ready to mount.
jonnybischof wrote: /edit:
From your first post: "215 x 215mm Ni-Cr Silicone self-adhesive heater" Ni-Cr is in fact an NTC material, but the resistance change is very small (a few % from 20 to 300°C) so I don't really see a good reason why your PSUs are failing...
Can you measure the actual resistance of the heater pad with a multimeter?
I will try to measure it this evening.

I have no idea why the PSU fails, as there should be enough safety in...
Roughly calc for the first setup,
config A with 12V/10A/120W PSU: working, but only 65°C
config B with 19,5V/16,9A/330W PSU: imediate shutdown, but...
1,5 Ohm of the NTC heater(12V/8A/96W) => 19,5V of the PSU => 13A/253,5W ==> the PSU data says 19,5V/16,9A/330W, so it should work... but do not...

Roughly calc for the second setup with new heater:
0,7 Ohm of the (?second?) heater (12V/16,6A/200W) ==> the PSU data says 12V/18A/253,5W, so it should also work... but also don´t...

I borrowed current measurement tweezers from my company and will try to get a measurement with it too.

Jonny, thank you for your advice - I know, this is a topic which periodically comes up by myself, and you try to give me a helping hand - THANK you buddy...
User avatar
jonnybischof
Reactions:
Posts: 310
Joined: February 15th, 2016, 2:48 am
Location: Zurich, Switzerland
3D Printer(s): UMO

Re: PSU issue... need a bit assistance...

Post by jonnybischof »

About temperature: You should add a layer of cork underneath the heater so that it can't radiate any energy downwards. That actually helps quite a bit.

I'm always happy to help (if I'm there...) ;)
User avatar
drayson
Reactions:
Posts: 254
Joined: February 15th, 2016, 5:04 am
Location: Graz, Austria
3D Printer(s): UMO with mods :-)

Re: PSU issue... need a bit assistance...

Post by drayson »

drayson wrote: Roughly calc for the second setup with new heater:
0,7 Ohm of the (?second?) heater (12V/16,6A/200W) ==> the PSU data says 12V/18A/216W, so it should also work... but also don´t...
Need to revise this data. Measured the resistance of the new heater - it has 1,2 Ohm !!
Roughly calculted, the heater needs 14,7A and produces ~180W with the supplied 12,24V... instead of the above mentioned...
So far so bad...
I tried also to hook up the PSU to mains first, let it "start" and then connected the heater - same result - imediate shutdown...

Next: Resistance measurement on the first setup...

//Edit:
drayson wrote: config A with 12V/10A/120W PSU: working, but only 65°C
config B with 19,5V/16,9A/330W PSU: imediate shutdown, but...
1,5 Ohm of the NTC heater(12V/8A/96W) => 19,5V of the PSU => 13A/253,5W ==> the PSU data says 19,5V/16,9A/330W, so it should work... but do not...
Revised the data of the first setup: Resistance measurement showed
- @ room temp.: 1,3 Ohm
- @ 40°C 1,3 Ohm
- @ 50°C 1,3 Ohm...

So:
config B with 19,5V/16,9A/330W PSU:
1,3 Ohm of the NTC heater(12V/9,3A/110W) => 19,5V of the PSU => 15A/292,5W ==> so it should work again, very close, but should...
drayson wrote: I borrowed current measurement tweezers from my company and will try to get a measurement with it too.
The current measurement showed nothing as both DELL PSU 's shut down too fast to get a reading... :-(

//Edit2:
Additional info found abount my 330W PSU: it can "provide 440W before it goes into protect mode and 22.5 amps before protect mode" so it´s even more strange...
User avatar
jonnybischof
Reactions:
Posts: 310
Joined: February 15th, 2016, 2:48 am
Location: Zurich, Switzerland
3D Printer(s): UMO

Re: PSU issue... need a bit assistance...

Post by jonnybischof »

That 440 W rating probably has a time constraint added to it? Like "it can provide 22.5 A for 1 second maximum" ?
That would be an usual thing to deal with inrush currents (when you power it up and have to fill empty capacitors which draw a huge amount of current for a very short time), and it also makes sense because the continuous current limit doesn't come from the components' constraints, but from the heatsinking which would be too weak to deal with that high current for a prolonged time.

Anyways. Summing it up:
From what you've told us so far, you're not doing anything wrong, and have tested the whole setup with multiple parts each. Still nothing works. That is... highly unlikely... :P

Have you tested everything using one and the same 3D printer controller?

There are two likely cases:
1. Some component is faulty (very unlikely since you have so many different combinations to rule that out)

2.
Image

Double and triple check the wiring. Make sure you have adequate wiring, make sure you have no GND loops. Maybe google up some drawings from other people and compare... You wouldn't believe it, but no electric circuit is too simple to make a mistake with it :P (my personal experience...)
Post Reply

Return to “Electronics”