@Sineos Just one question:
Why did you put pin 3 (clk) of the ADS1220 to ground? Can’t help but it looks like a short circuit to me.
Thanks for having a more detailed look. Every pair of eyes is valuable and I’m quite sure there is one or the other bug in it.
In this case, I think it is correct: CLK is used for external oscillators and should be put to GND if only the internal one is used.
CLK → GND = internal oscillator on 
Hi all around,
As I mentioned above, my plan is to have the ADS together with the ADXL running on one SPI bus.
Since I’ve seen you guys are working with KiCad, I put that on my disk and tinkered with it, until i got a schematic to show in principle what i have in mind.
Let me point out, that I’m aware this schematic is lacking everything needed to build up a proper circuitry. Nevertheless I’d like to ask if this can principally work.
Edit
@Sineos @garethky
I just like to encourage you specialists for integrating an ADXL345 chip on your board designs dedicated for mounting on printheads (clearly doesn’t make sense for the under bed mounted version).
As mentioned above, I’m just tinkering with the schematics. I’m afraid I’d never create my own PCB…
So I’m really looking forward for your projects - and try to influence you to fulfill my needs. 
Thanks in advance
2nd Edit
There is a thing to care for if the ADXL345 doesn’t have the SPI bus on its own:
So the schematics above tend not to work properly!
This likely will not happen anytime soon (at least from my side). I’m currently working on a board with a focus on the “low-noise” aspect and as it stands it has about double the size of something that would be tolerable as a tool-board. It could of course by shrunk quite a bit by using very small SMD that forbid hand soldering, but still I’m not sure whether it would finally match.
At the current stage we have too many blind spots regarding what is required in terms of:
- power supply quality
- filtering
- tolerable length of a load-cell wiring, i.e. having the board as close as possible to the load-cells vs. moving the board out and having longer load-cell wires
- influence of temperature on the quality of the readings
- etc.
I’m an absolute novice in such designs, so this will probably end up with more error than trial.
And, FWIW, I rather leave creating commercially available boards to companies who know how to do it. In the best case, our dabbling can serve as a source of ideas and proofs of concept.
Well,
i decided to start with the stuff i have at hand now and to make the concessions needed to go on:
- The Power supply will be taken from the GPIO of a CB1 Module, mounted on a Manta M8P Board. I cant offer a better one at the moment.
- I’ll use the noisy Breakout Board as is, since I haven’t seen any alternative suiting for mounting on printheads.
- So I’ll keep the length of the load-cell wiring very short (appx. 5 cm max).
- I’m using a full bridge, of homebrewn load-cells, simply strain gauges glued on a printed plastic part.
- The ADC will get mounted in immediate vicinity of the Hotend, so we get a hard test regarding to environmental influence (heat).
- The whole cabling between printhead and electronic compartment will be done with shielded cabling, including endstop and Fans, all in separate shielded cables - but excluding Heater - maybe thermal measurement becomes shielded together with the endstop.
- I’ll try the circuitry as shown in the above schematics with no filtering.
So loads of concessions here, but my aim is not to get an accuracy of one gram but to get just sufficient readings for endstop functionality.
After all I’m just curious if i get readings at all. If i do, we might have an idea, where the lower limit of functionality is reached - or missed.
Keep your fingers crossed!
1 Like
FYI, it is not valid to place an adxl345 directly on the same SPI bus as other devices, as the adxl345 may interfere with other devices (see the adxl345 documentation for details).
-Kevin
1 Like
Hi Kevin,
thanks a lot for your information!. That’ll save me from shipwrecking again.
Since this is my last blade of grass to get a reliable Z-homing in place, I’m very thankful you pointed me on this fact.
I’ve already made several attempts for Z-homing, but nothing turned out to work consitently:
-
CR-Touch is failing due to the high ambient temperatures i assume. The probe regularly fails to deploy at bed temperatures above 120°C.
-
Inductive Probe fails due to the fact that i have a pure Aluminum bed, covered by a thin sheet of Pertinax - no Steel to detect. The trigger distance is reduced below the thicknes of the Pertinax.
-
BD-Sensor is same as inductive Probe.
-
ADXL Tap Probing did not work out since my Printer is too rigid and too slow to detect the jolt.
So I’ll add another cable for a second SPI bus - not nice but no problem.
But how do I wire and configure the use of two SPI busses with my CB1?
This Story continues here.
I’m working on the code to enable the extra input configurations that you need.
Doing a whole toolhead board with an ADXL is still well above my pay grade. I’ve looked at the Huvud board and its very complex.
Thanks a lot.
That is quite not what I’m looking for. Running a TMC2209 stepper driver within an environment of 70°C is out of discussion, same is valid for high current MOSFETs. Having a few IO pins for fan, thermistor and endstop might be nice to have, but what is definitely missing is an ADC.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.