I’ve been printing with a cartridge of Durable for the last 1.5 months or so. 16 prints. According to the cartridge’s dashboard history, the sum of resin used for those 16 prints is 736mL. But the dashboard summary says this resin tank has only 276mL used with 725mL remaining. So with what went in to the tank at the initial fill, in reality the tank is all but empty. But if I relied on the Dashboard I’d think it was still 3/4s full.
The “accuracy” of this parameter is so poor, it’s all but worthless.
Doesn’t seem like the kind of bug that should be hard to fix. Since I can add the “Resin Used” data from all the prints that were logged against this cartridge and get a more accurate accounting, the Dashboard GUI ought to be able to do the same…
The really tough thing about this is calculating how much to charge a client based on resin used. Plus there needs to be an update or overhaul of preform itself to give us the ability to give accurate numbers when giving a client a base figure to charge them with. I can figure it out but with the overall cost of these machines it seems not such a hard thing to get right.
It’s clear the correct data is being gathered by the machine. There’s an accurate list of the prints I’ve run with the current cartridge, and if I click each one of those prints in the “Print History” list, it tells me how many mL were used by each print. If I add all the prints up, plus ~200mL for the initial fill of the resin tank, I end up with a number that seems very close to “reality” given the weight of the cartridge, I can tell it’s nearly empty and my manually generated sum comes to just a little shy of 1L. Yet in the very same Dashboard, under the “Details” tab, it says only 336mL has been consumed for the same set of prints.
Whether or not PreForm’s estimate of how much Resin will be used is accurate is a secondary question, though I think that calculation is probably fairly accurate. But the Dashboard isn’t adding up those PreForm estimates correctly.
It’s a simple bug, it has to be (I’m a software engineer) because I can easily do it more accurately by adding up the same data sources by hand.
What we do is… we know an empty cartridge weights ~400g (IIRC, I don’t have the # right in front of me), then we weight a full cartridge when it comes in and then again after each print. That way we know how much was used. It is a little bit of a pain but it is much more accurate then the calculation that the machine does.
Yeah. But my point is the calculation being done now appears to be fundamentally busted. The machine collects data for each print, including how much resin was consumed by the print, but that resin-used data does not seem to be incorporated in to the calculation itself. I don’t know what the resin-used calculation is based on, but it sure doesn’t look like it’s based on the actual info the machine clearly possesses. The resin remaining in the cartridge would be significantly more accurate if the calculation was being done properly, since the manual sum of the data that I did came up with something pretty close to the right number.
The calculation PreForm does seems to be fairly accurate. The problem is that the calculated number from Preform doesn’t appear to be what’s being used for the cartridge volume-used calculation. As described above, if I add up the calculated resin usage from Preform, from the “history” tab of the cartridge on dashboard, I get a number that suggests there is less than 100mL left in the cartridge, which is consistent with how the cartridge feels (weight and sloshy-ness test). But the Dashboard reports that the cartridge is still 66% full. The number displayed by the Dashboard is not using the numbers that Preform is calculating. It doesn’t appear to be using anything. I did a succession of prints, each of which used about 100mL, and the cartridge fill level didn’t change at all. In only just dropped to 66% on the very last print (when it should have been saying “0%” or something close to that).
The precision of the estimated cartilage level depends on the printed details - if the details have cavities or structures that can hold the resin, this trapped resin is adding on top of the material used for the printing.
The resin in the tank is also definitely taken into account when calculating the cartridge level - the printer won’t start printing until the minimum tank level was not reached. And here is something you need to take in account - the tank level sensor needs from time to time recalibration (it’s part of the firmware of Form3) . If the tank level sensor is not calibrated it may result in excessive amount of resin staying in the tank; technically this resin is not lost but can’t be used as the printer will stop printing if the tank level is too low.
There is also another reason for the big discrepancy between the estimated and real cartridge level - sometimes the cartridges are not fully filled. Personally I witnessed one such cartridge - while tank was filled, I printed details 100ml per print and instead of 10 prints I was able to make only 2 prints before getting empty cartridge. I contacted FormLabs and showed the Dashboard results and they agreed to send me a new cartridge. This replacement is only possible if you print one and the same details on regular base in big volumes, you use the Dashboard and get substantial difference between estimated and real cartridge levels.
To detect the semi filled cartridges I always measure them before putting into the printer - empty cartridge is ~320g so you should get ~`1320 in total. So far I see that the delivered cartridges are normally slightly overfilled (1010…1020g instead of 1000g)
But if I look at Print History for this same cartridge, and I click each print job listed and add up the mL that it says each job consumed, it adds up to about 930mL used.
So something in the resin-used calculation is fundamentally busted. It’s not using it’s own data to do the math. Dashboard thinks it is still 2/3rds full. Even though the print journal data for this same cartridge, added manually, says the cartridge is nearly empty.
If Dashboard did what I did, simply added up what Preform said the jobs required, it would be very close to the actual resin used.
I would add that none of the objects I have printed with this cartridge have internal voids that can capture resin. They all all open-sided objects that drain completely while they hang upside down on the build plate after the print completes. The only resin I’m carrying away is the skim coat that’s on everything at the end of a print.
And even if that was what was happening, it doesn’t explain the problem. It’s not that there is less resin in the cartridge than there is supposed to be, it’s that there’s a pissload more resin in the cartridge than there’s supposed to be.
Formlabs is worthless when it comes to fixing stuff like this. It’s doesn’t make them $ and I doubt it cost them any significant about of customers so they just ignore it.
