Share:
Share
Notifications
Clear all

Electrical Loads to calculate the cost of a print

peterb8f15b29
(@peterb8f15b29)
Famed Member

Hi,

I want to calculate the cost of my prints. There are 3 elements:

1. cost of filaments. This is easy as Cura will tell you the weight in grams.

2. The cost of electricity. I know the cost per kWh but can anyone tell me roughly the electrical load in watts for a Geeetech A10T. Geeetech cannot, they just say it is rated at 370W, but what is the running load when printing with a heated plate. Certainly the load will be high to get the hot end and bed up to temperature but after that it just tops up the temperature accounting for heat losses. Any thoughts?

3. My time to design the product which can be assessed for each item tp be printed.

 

Please help with Item 2.

Thanks

Peter

Quote
Topic starter Posted : 31/08/2021 1:37 pm
Luke
 Luke
(@lmf5000)
Expert Moderator

Hi Peter,

That's a great question. I've measured the consumption of my printers using an inline wattmeter like this 

I've taken the following measurements on my CR-10 mini, which is quite similar in size to your printer:

  • Printer electronics (excluding motors) - 10 Watts
  • Printer electronics + motors activated - 15 Watts
  • Nozzle heater - 30-40 watts warmup, ~20 watts to maintain 200-230°C
  • Bed heater - 230W when on, 0 when off, about 150W if your printer is equipped with PWM bed switching (else it just cycles between on and off every 30 seconds or so - my CR-10 mini was factory-fitted with an external MOSFET board to control the bed heater and I flashed a customized version of Marlin where I turned on bed PWM instead of using the default bang-bang control).

The peak consumption of the printer happens during warmup where it hits just over 300W. The average consumption of the printer (for energy or cost calculations) during printing is more like 200W. The biggest factor impacting this is bed temperature - at 50°C the bed has quite a low duty cycle (no hard numbers on this but I'd say it's only on 30% of the time), at 80°C the bed is nearing its limit and the duty cycle is almost 100% (i.e. heater on constantly).

Hope this answers your question adequately 🙂

 

Regards,
Luke

ReplyQuote
Posted : 31/08/2021 5:55 pm
peterb8f15b29
(@peterb8f15b29)
Famed Member

That's very useful Luke. The figures give are about what I would expect. I am not sure but I think my printer couls have PWM as the rate of temperature change reduces as it approaches the set point. Thank you for your help.

ReplyQuote
Topic starter Posted : 06/09/2021 4:02 am
peterb8f15b29
(@peterb8f15b29)
Famed Member

That's very useful Luke. The figures give are about what I would expect. I am not sure but I think my printer could have PWM as the rate of temperature change reduces as it approaches the set point. Thank you for your help.

ReplyQuote
Topic starter Posted : 06/09/2021 4:02 am
Luke
 Luke
(@lmf5000)
Expert Moderator

Hi Peter,

That might be PWM, but the way to check for sure is to measure the power consumption while the bed is approaching the setpoint. Even a non-PWM bed will use PID control (Proportional Integral Deriviative). In a nutshell it's a control algorithm where the output (in this case whether the bed heater is on or off) is dependent the input (in this case temperature) using a function with a proportional, an integral, and a deriviative component.

The proportional term is easy to understand - its output is proportionate to how far away you are from the setpoint. So when the bed is very cold it will command a lot of heating, when the bed is close to the setpoint it will command less heating, and when the bed is at the setpoint it will command zero heating.

The deriviative term is proportionate to the rate of change (how fast the measurement is changing). It's a refinement that helps avoid overshoots. For example, if your temperature gauge reading only updates once per second, and your printer notices temperature rising by 5 degrees per second, it would be prudent to cut the heating when you're 5 degrees below the setpoint, because the temperature will likely rise before the gauge registers the change (this is a bit of an oversimplification but you get the gist).

The integral term looks at the average over time and does gentle fine-tuning. For example if your setpoint is 60°C and your temperature is fluctuating between 60 and 62 °C (so average of 61°C) the integral term will slowly turn the heating down slightly until the average temperature has decreased to 60°C over a long period of time.

Because of all that, even with bang-bang control (not PWM), the printer will cut heating power before it reaches the set temperature, then wait for the temperature to stop rising (basically waiting for any residual heat to travel from the bed to the temperature probe, and for the probe to register the increase and send it to the computer), then it will turn the bed on and off as needed to stay within 1-2°C of the setpoint.

I know you probably already know this, but with PWM, the only difference is that instead of just on/off, the bed will be able to be turned on at a particular power level (say 25% or 75% power) instead of just completely on on or completely off. This is achieved by turning the power on and off extremely rapidly (hundreds or thousands of times per second) and varying the percentage of time it spends in the on and the off state (eg if the bed is on 50% of the time and it's a 230W heating element, it will consume 230W half the time and 0W the other half of the time, so on average the output power is 115W of heat, or 50% of the rated power). It's basically bang-bang control but on a timescale hundreds of times faster, and the benefit is smoother and more accurate temperature control. The drawback is that it's harder on the MOSFETS (electronic solid-state switches, aka transistors) controlling the bed, it creates more waste heat in the MOSFETS (since they heat up while they're changing state, but not when fully on or fully off), and it creates more electrical noise.

Hope this helps a little. I do recommend buying a power meter, I have quite a few around the house. If you want to go one better, you can buy a smart switch with power measurement, like a Sonoff POW-R2 or similar. It lets you turn the printer on and off remotely, and lets you track power consumption in real-time as well as keeping track of historical power consumption by calendar date, and costs about $25 :). For more simple devices, kill-a-watt make a unit that's popular in the US. But since I'm in Europe I buy these ones - https://www.ebay.ie/itm/194273233367?hash=item2d3b9649d7:g:xcMAAOSwa7Ngk87O

 

Regards,
Luke

ReplyQuote
Posted : 07/09/2021 10:49 pm
peterb8f15b29
(@peterb8f15b29)
Famed Member

@lmf5000

Hi Luke, following on from our conversation I have put together a spreadsheet to calculate the cost of a print.

It lists all the parts separately and then sums them for the total cost of the product.

It allows for multiple filaments used, even on the same part and allows the weight of the part to be entered either in grams or volume with the density of the filament.

The unit cost of the electricity is variable.

It is protected with a password (Io3DPPAd) if it is necessary to change the locked cells.

 

Please have a look at it and if you are happy with it, I am quite willing for it to be distributed to the Io3DP Members.

Regards

Peter

 

ReplyQuote
Topic starter Posted : 11/09/2021 4:54 am
Luke
 Luke
(@lmf5000)
Expert Moderator

@peterb8f15b29

Hi Peter, many thanks for this. I've forwarded it to the team for consideration 🙂

Regards,
Luke

ReplyQuote
Posted : 17/09/2021 7:41 pm
peterb8f15b29
(@peterb8f15b29)
Famed Member

Hi Luke,

I found a few minor errors but substantially correct. But I decided to  expand it for all materials and put them on a separate sheet. The updated spreadsheet is attached.

Kind regards

Peter

 

ReplyQuote
Topic starter Posted : 18/09/2021 4:59 am
peterb8f15b29
(@peterb8f15b29)
Famed Member

Hi Luke 

I also separated the filament volume from the weight (grams) as with very small parts it was taking the volume as grams. I have added Note 8 to only enter only one of them.

The spreadsheet has been updated to V2.1 as attached. Please disregard V2.

 

Regards

Peter

 

ReplyQuote
Topic starter Posted : 18/09/2021 5:11 am
peterb8f15b29
(@peterb8f15b29)
Famed Member

Hi Luke.

I see that there is the Training Webinarjam tomorrow and thought that the ability to cost your printed objects might be relevant so decided to speed up the proofing of the spreadsheet in advance of yours and your colleagues comments and thoughts.

I have attached Version 2.2 and attached my summary of the updated changes/corrections for some small anomalies that I found  in Version 2.1.

ReplyQuote
Topic starter Posted : 19/09/2021 8:31 am
Luke
 Luke
(@lmf5000)
Expert Moderator

Hi Peter,

Thanks again for this :). Sorry for the late reply. I made a few cosmetic and functional updates - conditional formatting for the "Material?" prompt, added a qty multiplier for the NPIs (unit cost x qty = subtotal), removed the currency symbol from the total and used default numbers that would look more reasonable for people using dollars, euros or pounds (basically electrical cost is about 0.13 and filament cost is about 30). Attached new version (same protection password as before). I will bring the new updates to the team's attention.

 

 

Regards,
Luke

ReplyQuote
Posted : 22/09/2021 8:38 pm
Share:
Share

About us

We’re a bunch of 3D printing and design nuts. We just want to make great 3D design available to everyone. 

Our mission

Good design takes time, great design takes a process. You can learn this with us, to build your skill set in this rapidly expanding market. 

Learn with us:

© Copyright: 2019 io3dp.com
Close Menu