Heated bed wont maintain more then ~100C

Hello everybody. I am not sure if this is the best place for this, sorry if it isn't. I have a Delta printer with a heated bed i purchased from amazon that is a 2Ohm resistance directly on the plate. I have been have adhesion problems with ABS (I have a 200mm wide role of Kapton on the way) and wanted to try to run a higher print bed temp(~110C). I am running my bed at 12V and it takes a long time to get up to temp even when I cover it in cloth. after i uncover it and it starts printing, it has a hard time maintaining over 100C, sometimes as low as 94C for prints where the fan is running 100%. I was wondering if my bed could handle upping the voltage and what others might have tried already. would it be better to just try to find a bed that will put out more wattage at 12V? I have plenty of overhead on my PSU (45A 12V line)

Correction: the bed is a 1.6Ohm unit listed here. 90W seems a bit low, I may look for a 15V supply that can put out about 10A.

Edited 1 time(s). Last edit at 03/26/2016 01:53AM by TbirdMan.

I started to think about it more, and i really don't like the idea of adding another entire power supply to my system. Has anyone tried a boost converter like this or similar? I think a boost converter run directly from the ATX 12V line controlled by another mosfet so I am not trying to run 12-15A through my RAMPS. the converter should give me the flexibility to tune my power as well.

One concern is the LEDs on the bed, would I need to desolder them? they don't really serve a function as i can't really see them.

I don't think you are looking at this the correct way.

A heater with 1.6ohms of resistance is going to pull 7.5A running on 12V (90W output). No power supply is going to make your heater pull more wattage. Higher voltage will simply lower current draw, but keeps the same wattage. Ohm's Law.

If you are using a computer supply, you need to verify if it is a single or double rail system. What power supply are you using? How are you powering your bed? Is it through the electronics controller, or SSR? If running through the electronics, is it using bang-bang or PID controller output? What is the max output setting if using PID? We need more specs man!

To me, it sounds like either bad wiring (are the wires getting hot when the bed is heating up?) or a bad connection. If it is either of these two, an electronics fire is in your future.

Edited 2 time(s). Last edit at 03/26/2016 05:16PM by MrBaz.

MrBaz, you're absolutely wrong regarding power and Ohm's law. Increasing the supply voltage will increase current and power, and therefore, heat.
The bed heater is a resistor. Ohm's law applies- current=voltage/resistance. Power=current^2 * resistance

TbirdMan, a dc-dc converter is a power supply. Semantics aside, you can increase power by increasing applied voltage via another power supply (or DC to DC converter), or get a heater with a lower resistance that will use more power and create more heat from the supply you have, assuming that supply can handle the load. Be sure that whatever switches power to the heater (MOSFET or SSR) can handle increased current and make sure the PCB can handle it too.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Ok, specs. My PSU is a 550W ATX power supply, one 12V 45A rail run through 8 wires (two for each + and - lead) to the RAMPS , so more then enough. As the digital dentist pointed out you are misunderstanding Ohm's law, doubling voltage for the same resistance quadruples power (P=V^2/R) and only doubles current (I=V/R) meaning it is more efficient. the bed is using PID that I ran the autotune on, max set to 255 (100%). PID constants are not the issue, 90W for 314cm^2 is just a bit anemic. it can hold around 100C printing if the part cooling fan is not at 100%, but barely. My bed is running through the RAMPS 1.4 MOSFET, the wires that came on the heater are beefy (maybe 10ga or 12ga) and get slightly warm to the touch under 100% dty cycle, but not excessively. the wires I added to reach the controller are either 18ga or 16ga (it is about impossible to read sometimes) from a computer power supply, doubled up on each line, they stay cool to the touch.

I realize a dc-dc converter is another power supply, I guess what i meant to say is I already have an ac-dc PSU that can provide more then enough 12V current, I would like to avoid adding another ac-dc supply as that would be both more inefficient, and would make it more difficult to implement the PS_ON control pin in the future. If i run 15V the current on the 15V side will only be about 9.3A, however the power on the 12V side will be around 12A, a bit more then i want to draw through the RAMPS board (especially with the polyfuses). I could either beef up the RAMPS with blade fuses and bridge wires to increase current capacity, or I can run a MOSFET from the RAMPS output to control a separate 12V line coming direct from the PSU. I like the second option more as it means if I should ever have to repalce my RAMPS board it should be a drop in affair rather then having to mod the new board.

I guess my biggest question is does anyone know what kind of current these MK3 direct on plate heaters can take?

Quote
the_digital_dentist
MrBaz, you're absolutely wrong regarding power and Ohm's law. Increasing the supply voltage will increase current and power, and therefore, heat.
The bed heater is a resistor. Ohm's law applies- current=voltage/resistance. Power=current^2 * resistance

TbirdMan, a dc-dc converter is a power supply. Semantics aside, you can increase power by increasing applied voltage via another power supply (or DC to DC converter), or get a heater with a lower resistance that will use more power and create more heat from the supply you have, assuming that supply can handle the load. Be sure that whatever switches power to the heater (MOSFET or SSR) can handle increased current and make sure the PCB can handle it too.

You are right in one aspect. I was mixing too many ideas in my head. What I was thinking is using a higher voltage power source allows you to run less current through similar devices rated for the same power output (if they are designed for it). Example: I have a 24V/250W heater for my heatbed and 24V allows me to run a touch over 10A. If I were to use a 12V heater, it would need twice the current to hit 250W.

That heater is designed for 90W. PERIOD. It is also marked for 12V. Using a higher voltage source will run at a higher wattage and possibly cause damage (read: FIRE). That 15V power source would make that heater output 140W. That is almost double its rated output. Suggesting anyone do so is foolish.

Edited 1 time(s). Last edit at 03/26/2016 10:06PM by MrBaz.

Quote
TbirdMan
Ok, specs. My PSU is a 550W ATX power supply, one 12V 45A rail run through 8 wires (two for each + and - lead) to the RAMPS , so more then enough. As the digital dentist pointed out you are misunderstanding Ohm's law, doubling voltage for the same resistance quadruples power (P=V^2/R) and only doubles current (I=V/R) meaning it is more efficient. the bed is using PID that I ran the autotune on, max set to 255 (100%). PID constants are not the issue, 90W for 314cm^2 is just a bit anemic. it can hold around 100C printing if the part cooling fan is not at 100%, but barely. My bed is running through the RAMPS 1.4 MOSFET, the wires that came on the heater are beefy (maybe 10ga or 12ga) and get slightly warm to the touch under 100% dty cycle, but not excessively. the wires I added to reach the controller are either 18ga or 16ga (it is about impossible to read sometimes) from a computer power supply, doubled up on each line, they stay cool to the touch.

I realize a dc-dc converter is another power supply, I guess what i meant to say is I already have an ac-dc PSU that can provide more then enough 12V current, I would like to avoid adding another ac-dc supply as that would be both more inefficient, and would make it more difficult to implement the PS_ON control pin in the future. If i run 15V the current on the 15V side will only be about 9.3A, however the power on the 12V side will be around 12A, a bit more then i want to draw through the RAMPS board (especially with the polyfuses). I could either beef up the RAMPS with blade fuses and bridge wires to increase current capacity, or I can run a MOSFET from the RAMPS output to control a separate 12V line coming direct from the PSU. I like the second option more as it means if I should ever have to repalce my RAMPS board it should be a drop in affair rather then having to mod the new board.

I guess my biggest question is does anyone know what kind of current these MK3 direct on plate heaters can take?

See my previous response for my messup on Ohm's law. PID settings ARE the problem. Turn off your autotune and set your PID settings back to defaults. I bet you your britches it will hit 110*C (that is if you don't have any bad wiring or connections).

it will hit 110 if I have the bed covered, but if the part cooling fan is on at all it plummets. I know how PID controllers work, I have programmed a few myself. that much error should have the PWM running 100% dty cycle. I turned off PID and went straight bang-bang and it took over 10min to go from 100 to 110. It is not PID, 90W is just under-powered for this size bed. I think the simplest solution is to find a higher wattage bed and either run it at 24V (meaning another ac-dc power supply, yuck) or beef up my ramps board bed circuit and run it at 12V. I suppose if I am getting a new heater anyway I could also go the SSR/120Vac silicone heater route.

Quote
TbirdMan
it will hit 110 if I have the bed covered, but if the part cooling fan is on at all it plummets. I know how PID controllers work, I have programmed a few myself. that much error should have the PWM running 100% dty cycle. I turned off PID and went straight bang-bang and it took over 10min to go from 100 to 110. It is not PID, 90W is just under-powered for this size bed. I think the simplest solution is to find a higher wattage bed and either run it at 24V (meaning another ac-dc power supply, yuck) or beef up my ramps board bed circuit and run it at 12V. I suppose if I am getting a new heater anyway I could also go the SSR/120Vac silicone heater route.

Then something is clearly wrong:

1 - What kind of monstrous fan do you have blowing on your heated bed? A box fan? Even 3 small 40mm part cooling fans are no way going to be able to dissipate that amount of heat.

2 - Why are you even using a part cooling fan with ABS? You shouldn't be running a fan with ABS at all (well, maybe if you are printing teeny tiny parts - at which only a bare draft should be required).

Even though 90W may seem quite small, it is enough. That 90W heater gives off 1.53W/sq.in. My 250W heater covers a lot more surface area, but still only puts out 1.67W/sq.in. Yes, it takes it ~15min to hit 105C (hottest I've ever needed), but it has no problem maintaining that temperature even in my modestly cold office.

it is a small 40mm blower with a custom duct I find it helps when bridging ABS. however when a heater like this is at it's limit it won't take much to cool it by a couple deg. I have been having warping issues and as part of the attempted solution I was trying to print at 110 bed temp. 100 worked ok, but still cooled a bit and got some excessive corner lifting. The advice I found suggested that temp's as high as 120-130 would not be out of line but the consensus I found was 110 for best results. It is not that I absolutely have to have 110deg, but it would be nice to have the option.

Also, looking at your printer, you have a glass bed with a silicone heater. mine is aluminum with the heater directly attached. the aluminum has a k value of 237 W/(mK) is going to dissipate more heat then glass at 1.05 W/(mK).

Quote
TbirdMan
it is a small 40mm blower with a custom duct I find it helps when bridging ABS. however when a heater like this is at it's limit it won't take much to cool it by a couple deg. I have been having warping issues and as part of the attempted solution I was trying to print at 110 bed temp. 100 worked ok, but still cooled a bit and got some excessive corner lifting. The advice I found suggested that temp's as high as 120-130 would not be out of line but the consensus I found was 110 for best results. It is not that I absolutely have to have 110deg, but it would be nice to have the option.

I'm in a similar position with abs and a mk2 heat bed. There's quite a bit of variation in the mk2 beds and I seem to have been fairly lucky in that it will hold 110 against 20C ambient, provided I don't have the part cooling fans on and the cold end fans don't spill air onto the bed. I'd quite like a little more power into the bed to get quicker heat up times - it's about 6 minutes to 95 but things slow down from there. However, I think the best approach to abs is to build an enclosure so that you can raise the enclosure temperature to about 50C. That will make life much easier for the bed and significantly reduce the thermal stress in the print.

I was quite impressed by that dc-dc converter you found, I didn't know you could get that sort of power at that price point. However, it would be under-rated for the heatbed, or at least running on the very ragged edge. There are some 20A models around, but it's starting to get more pricey. And if you do go down the route of boosting the power through the heatbed you will need to make sure everything in the pathway is well mounted, suitably rated and running at safe temperatures.

I think at this point my best and safest route will be to get a 120V silicone pad and put it on my original stainless steel bed, that avoids adding another power converter. even a cheap SSR can run a load like that all day and barely notice it.

PCB heaters are generally under powered. You get more heat from them by increasing supply voltage if possible (some supplies are adjustable). Bumping the supply up a few volts usually does the trick. If you don't have an adjustable supply you can try insulating the underside of the bed to reduce heat loss there but it's no substitute for increased power.

The real solution to the problem is to use a higher power rated heater that works at your available supply voltage. I run about 0.5 W/cm^2 on my 1/4" cast tooling plate bed and it gets to 105 C in about 5 minutes with no insulation on the underside of the bed. I suggest looking for a similar rated heater, but it depends on your willingness to wait. If 5 minutes is too long, get more power.

You can run a 12V heater from a 24V supply by limiting the PWM to a maximum value of around 64. That will limit the duty cycle and so the average power to 1/4 of what it would be when running on 24V continuously. If you crank up that duty cycle a little you'll get more power and get up to print temperature faster. Anyone who thinks it is unsafe is welcome not to use the duty cycle control that's built in to the firmware for this purpose.

You should have a fail-safe temperature sensor on the bed that cuts power in the event of an unsafe temperature rise regardless of the heater, rated power, and supply voltage.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I never liked the standard heatbed design with copper traces.
IMHO they just waste a lot of energy for very little heating.
With the power required to get a MK2 to over 90°C I can easily get a piece of steel glowing red in my induction heater.
In fact it uses less than 10amp to do this in a few seconds.
Despite some people claiming otherwise copper simply is not a conductor for this job as it's internal resistance is too low.
Designing a proper aluminum plate with build in heating is no problem at all and it with less energy it can get to over 100° with ease.
Think of hot plate with an evenly spaced heating coil running on 12 or 24V.
Don't have the means to machine a plate the size of my bed at home and don't have the money to have some prototypes done, otherwise I would be just laughing about bed problems.

Quote

IMHO they just waste a lot of energy for very little heating.

Not wishing to seem argumentative, but there isn't anywhere for 'waste' energy to go. Whatever power you put into the heatbed has to come out as heat, at least for DC supplies. If a pcb heatbed isn't heating up enough there's not enough energy going in. OK, I retract that a little, if you don't insulate under the bed you can waste heat that way, but that's true of any heating element.

The problem with getting a higher rated heater at my operating voltage is that to run ~157W (the power you suggest which seems about right) I would be pulling over 13A, well above my RAMPS board is rated to handle. I don't want to add another power supply, so I think a 120V heater + SSR is the simplest option. It requires the least modification to my electronics package.

You've just discovered one of the many problems with trying to minimize 3D printer cost by down-specing every aspect of the design and construction. You want more heat because the heater that came with the kit can't get the bed hot enough to print ABS (even though the ad for the kit said it had a bed heater and that it was ABS compatible) so you try to use a higher powered heater and you discover that the power supply is inadequate, and the PCB can't switch the extra current even if the power supply could deliver it. That's how they get printer kits down to $300.

Once you get the bed working right for ABS, you'll discover the next problem: if you don't enclose the machine and keep the print chamber warm (50C), you won't be printing anything with ABS that's more that a couple cm tall because the prints will warp and delaminate. That's something else they don't tell you before you buy the kit.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

To be fair, the kit didn't come with the heated bed, and the power supply it did come with didn't have the current for it. I purchased and mounted the heated bed and switched to an ATX PSU. I also have some plexi around here i plan to build an enclosure with. It was really my fault for not doing sufficient research before I bought the bed.

If you really need serious power get a dedicated switch mode power supply.
30amp often go for under 50 bucks on fleabuy.
These cheap ones might not have the cleanest DC coming out but to heat it bed that won't matter at all.
And one thing I noticed with standard ATX computer supplies is that they don't really loads changing from zero to max current.
In many cases a simple LC filter with a massive "C" is enough to make the ATX regulate more stable.

As for copper in the bed:
I won't go into the arguements about energy, heat and so on again.
If I ever feel a real need for a heated bed I will make one without copper for the heating and share the findings on it.
In the meantime just feel free to show me examples for low to moderate heating temps where copper is used - outside the 3D printing world

Copper isn't widely used for heating elements because it isn't a cost effective choice. That doesn't alter the basic physics of energy in == energy out.

How do you have the bed assembled? I have my Kapton heater sandwiched between an aluminum heat spreader and 1/2" MDF board. I then place a mirror on top of the aluminum. The only place for the heat to escape is through the glass build surface (heat losses creeping through the MDF board are negligible). The glass does not like to readily release heat to the air as compared to bare aluminum.

Edit: all my stuff is running on 24V.

Edited 1 time(s). Last edit at 03/28/2016 12:44PM by MrBaz.

my bed is a single unit, the heater is directly attached to the aluminum bed. I have a paper/bubble-wrap shipping bag wedged on the underside insulating the bottom. the top is just bare aluminum with masking tape. the bed is mounted to the chassis by three ~1cm brass standoffs, so I suppose it can lose a bit of heat there, but likely not much. You also have the benefit of the glass not conducting heat very fast, so what heat is lost to air will not be so quickly replaced at the surface to be lost again if that makes sense.

Quote
TbirdMan
my bed is a single unit, the heater is directly attached to the aluminum bed. I have a paper/bubble-wrap shipping bag wedged on the underside insulating the bottom. the top is just bare aluminum with masking tape. the bed is mounted to the chassis by three ~1cm brass standoffs, so I suppose it can lose a bit of heat there, but likely not much. You also have the benefit of the glass not conducting heat very fast, so what heat is lost to air will not be so quickly replaced at the surface to be lost again if that makes sense.

Yep. Makes sense. What if you were to place a piece of glass on top of your aluminum?

might be worth a try.

The easiest and cheapest way to get around the problem is to bump up your power supply voltage to around 14 volts, and add an automotive 30 amp relay to do the switching instead of putting full current through the polyfuse. I've done this to two printers, and it works well.

Quote
nebbian
The easiest and cheapest way to get around the problem is to bump up your power supply voltage to around 14 volts, and add an automotive 30 amp relay to do the switching instead of putting full current through the polyfuse. I've done this to two printers, and it works well.

that works if you have an adjustable power supply. I am using an ATX power supply and have no desire to add another supply. I am going to try the glass bed surface, but if that doesn't work I am gonna go with a 120V silicone pad heater and SSR.

I'm not trying to de-rail this thread (OP, let me know if this makes you mad) -- but I'm actually having the same problem, but with my Prusa i3. I'm trying to print my first print, I've turned it on, and believe I have it calibrated and ready to go. I've initiated the print, but the heat bed is taking an awfully long time to heat up (to the suggested 110 degrees)... is that high of a temp just too much for the cheap Prusa i3 (it's the setting the manual says to use, although I'm aware the manual for this model isn't amazing)..

Thanks for any input, and Moderators please delete if this is incorrect... it's just that I have the same problem

I generally print ABS with a bed temp of 100C - I very rarely use 110C. This is with a Mk3 ali heatbed, insulated underneath.

Just a note, I'm using a 35 amp PS with an SSR and have bumped up the output from 12 volt ti 14.5 volt. After loosing 1.2 volts to the SSR I'm left with 13.3 volts to the heater. That and placing a piece of cork between the lower and upper plates and some peel and stick window insulation around the edges (and in between the plates) my temps went from struggling to reach 110c to 110c in 15 minutes. The insulation foam actually seals the gap between plates and really helps.