RE: PT100 with amp board
Ignore the values for gain etc above.
Of all the chip schematics in the world for the amplifier chip, I go and find the one with the wrong values in the gain calculation!, I thought things seemed a bit off.
Right, The gain is actually 10 (9.998179 to be precise if components are perfect)
I would try replacing R5 from 5k49 value to 3k3 to give a gain of 16 this would give outputs as follows:
With 5k49 which gives G=10
0 Deg C Vout = 1.110909 410 Deg C Vout = 2.692565 a difference of 1.581656 V
With 3k3 which gives G=16
0 Deg C Vout = 1.774411 410 Deg C Vout = 4.300728 a difference of 2.526317 V
This would give a theoretical maximum readable value of 539 Deg C, Way hotter than we print at.
A new table is easy to create for theoretical output, but a real life lookup table is better since PT100 sensors are not linear, but the values could also be inferred from existing output tables.
Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet.
RE: PT100 with amp board
Yes, but you're intrepid enough to go find the schematics and run the SPICE simulation. 😀
Off topic, but after looking at how small and close the resistor is to the jack, I was not satisified with the pencil on my rework station. Jack is also too close for my hot air pencil. Real nano-tweezers are out of my hobby price range. Picked up one of the TS100's the RC guys have been raving about. Tried it out and instantly fell in love with it. Battery operation capable, heats up super fast, and SO maneuverable. Had fun using it to rapidly knock some SMD off old boards. Should have gotten one ages ago.
RE: PT100 with amp board
It is tiny, in't it, I know I would not like to try removing ithe thermistor input pull up resistor either, maybe 30 years ago, but then components were HUGE in comparison to today.
I'm glad I have a Mk3, but I am planning on upgrading my printer to a PT100 so the modifications to the PT100 amplifier board may just get done anyway to improve precision.
Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet.
RE: PT100 with amp board
Yes, but you're intrepid enough to go find the schematics and run the SPICE simulation. 😀
Off topic, but after looking at how small and close the resistor is to the jack, I was not satisified with the pencil on my rework station. Jack is also too close for my hot air pencil. Real nano-tweezers are out of my hobby price range. Picked up one of the TS100's the RC guys have been raving about. Tried it out and instantly fell in love with it. Battery operation capable, heats up super fast, and SO maneuverable. Had fun using it to rapidly knock some SMD off old boards. Should have gotten one ages ago.
You don't need hot microtweezers do remove tiny SMD's. Simply placing a blob of solder over both contacts of the tiny resistor, so that both pads heat up, and then removing it quickly with a needle or tweezers works just fine. Now putting these tiny SMD's back is harder, but since you can do one pad at a time, it's still possible.
On the other hand, even with the limited voltage swing the resolution might be enough - the ATMega2560 has 10-bit ADCs with selectable references and more can be achieved using oversampling and decimation. So I believe that a pure software solution should be possible, too.
Since the resistor network on the Einsy basically works as a /2 divider, the swing at the ADC pin would be 0.75V without hardware modifications of the amplifier or the Einsy. With a default config of the ADC that'd give a resolution of 2.66 °C, which would work, but is not great.
I'll need to take a look at how it's configured for the thermistor ...
RE: PT100 with amp board
I used pins 2 and 5 on the top row (one of them is the same input for mmu2) and one wire to the thermister input. The picture is a old setup and I have the amp away from anything that would cause interference
Been running with a pt-100 (with the amp) for about 2 years now without a issue.
Let me know if this helps 🙂
RE: PT100 with amp board
Can't quite make out which pin and header from your pict
RE: PT100 with amp board
It's the same header the mmu2 and filament sensor (bottom row of pins) except we use the top row. Pins #2 (if you look at the mmu2 it is not being used) and pins #5 (counting left to right) on the top row. If you are using the mmu2 with a pt-100 you have to share that # 5 pin. Although probably not safe, I shared that #5 pin at one time with both and it worked via a y cable I made.
There was a big thread on this with the mk2, but for whatever reason I can't find it 😔
RE: PT100 with amp board
The purple wired pictured goes back to the thermistor input header... It doesn't matter which pin I don't think.
RE: PT100 with amp board
Dobbewan, thanks. So you are tapping Gnd and VCC from the top row of the header to get power and GND for the amp.
Signal back from amp is hooked directly to your Thermistor input header. (I'm pretty sure the pin DOES matter there because one is GND, while the other is signal input)
What is MOST interesting is that you are feeding the thermistor input without doing anything to the resistor network that we have been discussing as needing modification. You have been getting accurate temperature data despite the 4.7 K pull up resistor!? I guess only way to verify is 2nd calibrated thermistor to read temp simultaneously.
RE: PT100 with amp board
Correct. Like I mentioned, there was a very detailed thread on this when it was first added to the firmware in the mk2 hayday. I would give credit to those involved in importing the tempature tables, but for whatever reason since the new forum format, it is gone.
Forgive me, but yes that thermistor pin does matter. In my picture I have it on the right thermistor pin. Lately I have been traveling, so I unfortunately don't have my printer or notes with me. My goof.
As far as accuracy...
It is very accurate actually. In the old thread that I can't find, they actually test them side by side with a thermistor. The only issue I have is colder tempatures. Anything below 200c it seems to bounce around... But it could be due to the 60w overkill heater I have. I was having crazy variences for a bit, but found out the stock molex plug was the culprit and replaced with a xt-30.
RE: PT100 with amp board
One of the guys on the thread even added one to the heatbed... Although not needed, he claimed it was more accurate.
RE: PT100 with amp board
Still surprised the thermistor input can be driven that way. Well that's specs vs real world for you.
RE: PT100 with amp board
Specs vs real world always gets me buddy 🙂 I always look at work designs on paper and think "no way I can make that work" Pretty common with my occupation
Side note : Now that I have everything back together on my build, hopefully I can add more input to the vfa thread with the steppers!
I do remember in that thread that without the e3d amp the temperature is off with a varience of 10c or so. So the amp is highly recommended.
Things I have done to improve everything if you buy the pt-100 kit with amp...
- ditched the long provided connecter wire to the einsy. It is not very good wire for one and it is not the most flexible. I opted for silicone jacketed copper wire.
- Cut the molex off the pt-100 and soldered a xt-30. The stock molex is crimped and ends up right at the start of the wire holder on the back of the carriage. This causes the wire to pinch inside the molex and lose connection. I think this is due to the fact that the connection pins are crimped and on the single small copper wire on the pt-100 it bends it back and forth until the crimp digs in and eventually loses conductivity . I literally was going crazy trying to chase down that issue for about a week. Soldered is always better than crimped in my opinion anyway.
-placed the amp a good proximity from the einsy to avoid any interference (don't know if this improved anything, just piece of mind)
So far, in 2 years those have been my improvements. Anything to make it more accurate/stable helps 🙂
RE: PT100 with amp board
Attached is a picture of my wire loom/connectors for the 60w heater and the pt-100. Trying to make everything as serviceable as possible with disconnects on everything 😊
RE: PT100 with amp board
I like your separate disconnects. I've contemplated doing that several times, but got too invested in making it a unified connector. Your separate connector solution would make servicing individual components easier.
RE: PT100 with amp board
Thank you !
I will eventually have a unified connector or connectors on the einsy end before it exits the enclosure. For right now, it makes it easy to drop the extruder assembly and take the whole frame out of the enclosure for service. I used 4 pin xlr's for the z and y steppers and couldn't be happier. Soon I will have base mount xlr hub instead of the ugly male connectors 😊
RE: PT100 with amp board
Slightly different, but how about this thermistor from Slice engineering, available in the UK from here: https://www.3djake.uk/slice-engineering/mosquito-thermistor
It is rated up to 450 Deg C and would also need a mod to firmware, but no hardware modifications required.
Thermistor table: https://docs.google.com/spreadsheets/d/e/2PACX-1vTHfmSmnGNEuHzjCHEjkA9_DUgnCD9z5E_7Gk1BjUgBYEBtJ6Dlj3hhnFuhFeOfnIc_7zE69VS1tAbx/pubhtml
Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet.
RE: PT100 with amp board
Slightly different, but how about this thermistor from Slice engineering, available in the UK from here: https://www.3djake.uk/slice-engineering/mosquito-thermistor
It is rated up to 450 Deg C and would also need a mod to firmware, but no hardware modifications required.
I dunno if this would be any easier than the PT-100 and the amp. No real big hardware modification is done with the PT-100+amp but adding 3 female jumper wires in the correct pins. Plus adding the Pt-100 in the firmware is as easy as uncommenting a line in your configuration_prusa.h Seems like the pt-100 would be less hassle and more accurate according to the temp tables in thermistortables.h in the firmware.
I am curious though.... I would like to play around with this, but you would have to import the temp tables and compile and I am honestly unsure as to how you would exactly do that without testing (since nobody has done this yet as far as I know) and comparing the Pt-100 and the stock.
I was honestly going to just get the mosquito and add my 60w E3d heater and pt-100 to it and see how well that worked since I have heard nothing but good things about Slice..... Just sucks... I have my E3d V6 all decked out, and I don't know if I want to dump anymore money into it since I have the bondtech geared extruder body
RE: PT100 with amp board
The e3d heater and pt100 should fit the mosquito as they are the same size.
You may want to invest in some boron nitride paste though.
On the e3d there is a screw which pushes the thermistor body against the hotend wall in the hole for it, but there is nothing like this on the mosquito, just two screws at the ends to stop the thing falling out, it relies on the boron nitride paste, packing the space between the thermistor and sides of the hole, without this the thermistor will rattle around in the hole and not make a good contact with the sides, the same goes for the heater cartridge, it is not clamped as such, just prevented from falling out the sides. You may have an issue with the 60w e3d heater due to the 90 deg exit of the wiring and the screw that holds the heater in from the end on the mosquito.
Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet.
RE: PT100 with amp board
The e3d heater and pt100 should fit the mosquito as they are the same size.
You may want to invest in some boron nitride paste though.
On the e3d there is a screw which pushes the thermistor body against the hotend wall in the hole for it, but there is nothing like this on the mosquito, just two screws at the ends to stop the thing falling out, it relies on the boron nitride paste, packing the space between the thermistor and sides of the hole, without this the thermistor will rattle around in the hole and not make a good contact with the sides, the same goes for the heater cartridge, it is not clamped as such, just prevented from falling out the sides. You may have an issue with the 60w e3d heater due to the 90 deg exit of the wiring and the screw that holds the heater in from the end on the mosquito.
Nice!
I actually have boron nitride paste handy 🙂
Thanks for the info! I am really interested, but I am unsure of how great the improvement can be over my E3d gold with titanium heatbreak and nickel plated/nozzle x nozzle would be for the stuff I print. I usually print alot of 3dx CF/PC, PPC and PEKK on my main machine and have it pretty fine tuned for it, but I didn't know if anybody else was pushing the Mosquito to those temps? The highest I can print on my E3d is about 370C, but that is right before the bondtech SLS nylon starts to soften. No real limitations on the e3d so far from what I can tell and I don't plan on ever getting to that 450c mark 🙂
I am curious to know the difference in the E3d heater vs the Slice heater though. Looking closer, it does seem the angle is going to be a issue. I may just order a mosquito and see what happens after I get my other prusa up and running.
Thanks again!