Hot end maximum wattage possible



  • Hello,
    I have a printer we designed having two extruders, EACH have TWO heaters x 50 watts x 12V.
    therefore, each hotend has 12V x 100watts which equals to 8.33 Amps load.
    I am changing over from a ramps setup. I bought the DuetWifi v1.03 board. Will this handle this power or would I need to power these through a solid state relays?
    I do not know if I am right, but somewhere I saw that the maximum power that the board can take in each hotend terminal is 5 Amp.
    Could somebody clarify please. Thanks



  • "A 2-pin 3.5mm terminal block is provided for connecting each of the two supported hot end heaters.

    The design is rated to 5A per extruder heater connection"

    https://duet3d.dozuki.com/Wiki/Connecting_extruder_heaters

    "From Version 1.04 onwards the Duet 2 is fitted with 3 blade fuses:

    • 1A for the VIN going to the fans
    • 7.5A for the VIN going to the heaters and stepper motors
    • 15A for the VIN going to the Bed heater."

    https://duet3d.dozuki.com/Wiki/Hardware_Overview

    MOSFETS/SSR will be the way to go. With that sort of heating capacity read up on heater fault detection. Incorporating a mechanical relay controlled by PS_ON to kill the feed to the heaters (not fans) would be wise.



  • Rebuilding the machine for 24V would half your current requirements and reduce the thickness of wire needed.

    The on board mosfets would cope, but you would blow the fuse. Feeding the heaters externally and usung the switched ground on the board would work.

    Would still be wise to use a relay to break the feed to the high side of the heaters.


  • administrators

    @doctrucker said in HOT END MAXIMUM WATTAGE POSSIBLE:

    Rebuilding the machine for 24V would half your current requirements and reduce the thickness of wire needed.

    That is my recommendation too.

    Do the heaters need to be as much as 50W each?

    The mosfets and terminal blocks can pass well over 5A safely, probably 10A. It's probably safe to increase the fuse rating on the 1.04 board to 10A but I wouldn't go any higher than that.



  • Thanks Guys,
    that is what I principled in mind, but sharing ideas and getting people to say what they think is key to avoiding a trap. I tried using larger nozzles and found that my heater was unable to cope and if you go for materials requiring higher temperatures in my opinion would be necessary- I think. If the printer runs fast, it would need to melt instantly.
    correct me if I am wrong.



  • @doctrucker I think I will go that way. In fact, having a large heatbed, I had to arrange a 24V Ac 400KVA transformer with a solid state relay already.



  • From the work that I've done with large diameter nozzles, a bigger heater doesn't help. There is no getting away from the fact that it takes time for the heat to find it's way from the outside surface of the filament through to the core, especially as filament itself is y poor conducter. I played around with an 80 watt heater which didn't help with the melt rate but was a complete PITA to control.


  • administrators

    @mendelevium said in Hot end maximum wattage possible:

    @doctrucker I think I will go that way. In fact, having a large heatbed, I had to arrange a 24V Ac 400KVA transformer with a solid state relay already.

    The other option is an AC mains voltage bed heater, with appropriate safety precautions. See https://duet3d.dozuki.com/Wiki/Choosing_a_bed_heater#Section_Bed_heater_driven_using_a_Solid_State_Relay.

    From the work that I've done with large diameter nozzles, a bigger heater doesn't help. There is no getting away from the fact that it takes time for the heat to find it's way from the outside surface of the filament through to the core, especially as filament itself is y poor conducter.

    True; but if you can solve the problem if getting the heat into the filament (e.g. by using a longer melt zone) , then you do need more heating power. At least, this is what E3D told me was limiting their prototype Super Volcano hot end (i.e. insufficient heating power to maintain temperature at high extrusion rates).



  • @dc42 said in Hot end maximum wattage possible:

    From the work that I've done with large diameter nozzles, a bigger heater doesn't help. There is no getting away from the fact that it takes time for the heat to find it's way from the outside surface of the filament through to the core, especially as filament itself is y poor conducter.

    True; but if you can solve the problem if getting the heat into the filament (e.g. by using a longer melt zone) , then you do need more heating power. At least, this is what E3D told me was limiting their prototype Super Volcano hot end (i.e. insufficient heating power to maintain temperature at high extrusion rates).

    I solved the issue of getting heat into the filament by using multiple melt chambers radially located around a central heater cartridge. Managed to get up to 300mm/sec with a 0.5mm nozzle and 0.3mm layer height before running into problems. But there was no drop in temperature and that was with a 40Watt heater https://somei3deas.wordpress.com/2017/06/22/exploration-of-print-speeds-with-a-diamond-hot-end/

    That was with 3 melt chambers. The 5 colour (melt chambers) is on my list of things to test along with larger nozzles.

    When i tried an 80Watt cartridge, the only way I could control the wild temperature swings was to reduce the PWM to 50%, effectively reducing the heater output to 40 Watt.



  • Basically, i have designed a heatblock which consists of two cartridge heaters adjcent to the nozzle. It is in aluminium and very efficient. I intend to either produce it in brass or better still in copper. I have not yet tested the printer`s extruder capability to melt material fast. I believe that the heating must provide enough energy to fuse enough plastic otherwise speed serves no purpose. Secondly my idea is to strenghten the extruding capability by a geared arrangement with metal gears. I do not like extruder motors mounted on carriages for the inertia they cause during movement, so it will be bowden controlled. Eagerly waiting for my duet wifi to turn up and try it out...



  • Hmmm if you have Two 12V heaters you can put them in series and power from 24V, this would 'reduce' the current going through Duet.



  • @dragonn would they provide the same power? I would think half as much power. Not a bad idea to run on 24v but i think that would halve the heating power...



  • @mendelevium No, when you run two 12V heaters in series from a 24V source the power will stay the same. Ohms law 😄



  • ..........until one heater fails and you end up with 24v across a single 12v heater........



  • @deckingman yep very bad idea...



  • @deckingman I guess you mean fails shorted? Haven't seen that. In that case, what happens when a single 12v heater on 12v 'fails'?

    If you have fusing in place to prevent a failed 12v heater on 12v from doing bad things, that same fusing would protect a 12v heater on 24v (as current would be doubled).

    And if you don't have the fusing, well, I'll take a 2x jump in current over a 10x jump in current.



  • @elmoret I wasn't thinking so much about blowing fuses, more about what happens when you suddenly lose half the resistance. The one remaining heater is going to get bloody hot I'd have thought, although in theory the PID should control it to some extent and prevent a fire - hopefully, maybe?



  • @deckingman said in Hot end maximum wattage possible:

    The one remaining heater is going to get bloody hot

    How hot would a shorted single heater get? Especially when considering that not all heater shorts are guaranteed to blow a fuse or a trace on the PCB (could have a partially shorted heater resulting in say, half the normal resistance).

    I guess I just don't see how its any more dangerous than a single heater setup.



  • Firstly, incase you are not aware fusing is to protect wiring, not devices. Typically you would calculate or measure expected draw, then find the next suitable fuse rating above this. Wiring is then sized so that the chosen fuse level is no more than 80% of the rated current of the wire. Further increases in wiring cross sectional area may be required to avoid unacceptable voltage drops. Where you have a more complicated system the wiring to the specific device maybe larger still, just to save the need for many different cross sectional area wires in different colours from the bill of materials.

    If a single heater fails short the current multiplies many times and will blow a fuse pretty quick as it blows to protect the wiring.

    The fail short of one 12V heater of two in series would cause the resistance to half, doubling the current. This will double the rated power of the heater and things will get smokey fast. Much like accidentally leaving a voltage select switch on a computer psu at 110vac...

    This would be difficult to protect against with fusing as a fuses are often only guaranteed to quick fail at double rated current. Time to fail increases as you reduce the current towards the fuses rated current, at which the time to fail is somewhere between anytime and never. Fusing is also only intended to protect the wire, not the device.

    Wiring would have to be sized to be safe for the expected current through one 12V heater when supplied with 24V, double the expected current through two 12V heaters in series.

    A current chopper style circuit could make it safer, as after all most of our stepper motors rated voltages are much below drive voltage.

    Easier to use a 24V heater! We're talking about less than a £10 here, how many hours work would it be to design and test a current chopper that never chops in normal use and then would need routine testing?



  • This post is deleted!


  • @elmoret said in Hot end maximum wattage possible:

    @doctrucker said in Hot end maximum wattage possible:
    Well, that's just not true, or there wouldn't be fuses on printed circuit boards. Or in plugs. You can use a fuse to protect a wire or a device. But don't take my word for it!

    Plugs often have wires coming out of them. Printed circuit boards have wires coming out of them, and their traces are un-insulated wires. Fuses are not quick enough to protect the likes of the mosfets on the board which you are proposing the fuse protects in the fail-short failure mode.

    Why? Why wouldn't the fuse blow? No sense in fitting a fuse 2x that of the rated load.

    A typical guaranteed quick/near instant blow is twice the fuses rated current. If you've used cabling whose rated current is above the fuse then the cabling will take the brief short to earth for long enough to blow the fuse. To avoid nusance trips you would have to size the fuse above the normal current to avoid nusance trips, which is going to mean double current is not twice your fuses rated current. Your margins for fuse selection are getting small, and the range of fuses is not infinate - hence why there is a 15A fuse on the duet 2 for bed heater not 18A.

    This would be difficult to protect against with fusing as a fuses are often only guaranteed to quick fail at double rated current.

    Which would be experienced in this case.

    No it wouldn't if you didn't want nusance trips by selecting your fuse at exactly expected draw.

    Fusing is also only intended to protect the wire, not the device.

    That is again, false. Install a fuse that is rated for ~10% more than the rated load. In OP's case, I'd use a 10 amp fuse. A 10 amp fuse exposed to 16A (resulting from one of the heaters dead shorting) will blow in under 3 seconds. I doubt you'll even be able to detect a noticeable change in temperature in that time.

    This is where the OP may accept a reduction in safety magin if they wish, but my original suggestion of SSR/external mosfet is safer. Besides, at 10A you've aready gone above spec for the input.

    A current chopper style circuit could make it safer, as after all most of our stepper motors rated voltages are much below drive voltage.

    That's what a PID loop is.

    I think you're wrong here. PID is a control method, not a means to directly control or limit current and has no way to detect an over current. I was referring to active current chopping like in the stepper motor drivers. It would open circuit if the current exceeded a set limit but not reset. Much like the system in the Tesla battery pack which blows a pyro disconnect in the event of overcurrent.

    Easier to use a 24V heater! We're talking about less than a £10 here

    He needs 100w. 100w heater cartridges are not easy to find, plus they result in a pretty extreme heat density.

    Refer to my original commment regarding external ssr/mosfet. Could also do 3 40W cartridges which at a 100% duty will pull 5A so safe on the Duet spec. If you don't want 120W of heat adjust the max duty down.

    Edit: Just re-read dc42's earlier post that said the inputs can pass over 5A safely and probably 10A. Looks like the board could take the fault current of one of the two heaters shorting. I still would prefer external switching and 12V over on board and two 12v heaters running in series from 24vdc.

    Finding suitable 24V heaters or multiples thereof would be better still.



  • @doctrucker said in Hot end maximum wattage possible:

    @elmoret said in Hot end maximum wattage possible:

    @doctrucker said in Hot end maximum wattage possible:
    Well, that's just not true, or there wouldn't be fuses on printed circuit boards. Or in plugs. You can use a fuse to protect a wire or a device. But don't take my word for it!

    Plugs often have wires coming out of them. Printed circuit boards have wires coming out of them, and their traces are un-insulated wires. Fuses are not quick enough to protect the likes of the mosfets on the board which you are proposing the fuse protects in the fail-short failure mode.

    As for the FET, the Duet boards use IPD036N04L, which are rated at 90A continuous. It'll handle 200A for 1ms. The fuse is gonna blow before the FET, look at the time response curves for fuses and compare to the FET.

    This would be difficult to protect against with fusing as a fuses are often only guaranteed to quick fail at double rated current.

    Which would be experienced in this case.

    No it wouldn't if you didn't want nusance trips by selecting your fuse at exactly expected draw.

    I agree! You should oversize by 10-15%. Covered this in my reply.

    Much like the system in the Tesla battery pack which blows a pyro disconnect in the event of overcurrent.

    Sounds like a fuse!

    I don't wish to argue further, but hopefully this helps: have you ever looked at how a heater cartridge is made? It is a wire wrapped around a core, inserted into a metal cylinder. It could short in a variety of ways, resulting in resistances from nominal down to a few tenths of a ohm.

    Knowing that, what is the realistic difference between that, and two heater cartridges in series? Put another way, what is the difference between two 20mm long heater cartridges in series, and one 40mm long heater cartridge? (nothing, aside from physical packaging)



  • I'm a (retired) mechanical engineer, not an electronics expert but I've read all the posts above about FETs and fuses, and still running two 12v heaters in series from a 24v supply just seems an unsafe thing to do, unless I've missed something somewhere.

    Just about every electrical fault I've ever had with my printer has been some sort of wiring issue (bad crimps, broken wires, faulty terminals) and almost never a component failure.

    So in simple terms, if one used two 24v heaters in parallel and a wire came adrift on one of them and shorted to ground, it'd take out a fuse. That's providing you have a V1.04 board or later - mine is an early one without fuses AFAIK, so something would go pop instead of a fuse but no matter, it would still fail safe (more or less). But if one used two 12v heaters in series, and the same fault occurred on one of them, then it wouldn't be a short to ground which would take out a fuse or some other component, but a halving of the resistance and so the other heater would get very hot, very quickly and potentially be a fire hazard would it not? Or maybe even worse, the PID control would catch it and the user might never know until the MOSFET failed. In that case it would be fed continuous 24 volts, the firmware would detect the sudden rise in temperature but would be unable to turn it off because of the way the MOSFET has failed (or at least that's how I read it here https://duet3d.dozuki.com/Wiki/FireSafety).

    I'm just going by the fact that when I tuned an 80Watt heater, the firmware gave me a warning that it was overpowered and potentially a dangerous fire hazard. I can't help thinking that a 12v 40 watt heater being fed 24v would in effect turn it into an 80Watt heater.

    Maybe I've just missed something in all the above discussion, but my gut feel is that it's an unsafe thing to do.



  • @deckingman As I've tried to explain a few times now, there is no guarantee that a heater cartridge exists only in a full short or a no short situation. It can also be partial shorted.

    As such, running two 12v heater cartridges in series no more "dangerous" than experiencing a partial short on a single heater cartridge.

    If you are running a single heater cartridge, and the bottom half of it shorts, how is that any different from the scenario you're fearful of? The other half of the heater cartridge would get very hot, very quickly (until the PID loop kicked in), just like the scenario you're concerned about.

    And we all run single heater cartridges without worry.

    And also, there's no need for a v1.04 to have a fuse in the hotend circuit. One can add any fuse they wish, to any board!

    I need to find a way to disable notifications on this thread before I lose my sanity. 🙂



  • @elmoret Yes, I hear what you are saying and I take your point but ...................well let's just agree to differ before we both lose our sanity ☺


Log in to reply