Welcome to the SeeMeCNC Community

Xenocrates wrote:I support replacing the I2C with say, RSS485, run over a quarter mile spool of rusty barbed wire before it goes to the Rambo (I kid, but there are far more resilient standards, and something like RS485 is high enough data rate even at over a kilometer of length). I2C is a fine protocol for small runs, and in it's intended use case. On a board at the end of a multiple foot wire where it's next to a very electromagnetically noisy heater is not a use case that makes ANY sense for it. Admittedly, it's easy to implement, and there is a specific 4 pin connector on the full size rambo for it.

I2C was never meant for this kind of application. It literally means Inter-Integrated chip. I standard for talking between chips on a single board in close proximity.

One way to get past the length of the cable is simply to put the micro on the same PCB with the accelerometer, or really close to it on a separate PCB. In this case, I think the idea is to plug the CPU into the connector on the HE280 and use and endstop for the probe signal.

I suggest going one step further and include a spot on the PCB for an accelerometer to be populated. That way, the same board could be used for those without the HE280 by simply populating the part.

Also, there's no reason you can't have several screw patterns on the PCB so that it will mount to different hot ends by using the appropriate set of holes.

mhackney wrote:Others have thought of that but I'm not sure anyone has tried an acoustic approach. But, putting materials like PEI on the glass changes the properties and would need to be taken into account. And "loud" is relative. The noise from movement would need to be taken into account. Much more work can and should be done on probes!

In fact RRF already includes some code to support an acoustic approach. If you select Z probe type 6 (or type 7 in the latest 1.16 beta version), then the motors will vibrate the effector up and down while probing. A microphone mounted on one of the towers or a piezo sensor on the effector picks up the sound and either the main firmware or a separate microcontroller looks for an increase in the second harmonic when the nozzle contacts the bed.

The main disadvantage with this approach us that the vibration frequency is limited by the mass of the system, and that limits the probing speed because you need something like 4 cycles to detect the increase in the second harmonic.

I am just starting to test a simple solution for the V3 to interface with a Duet. In parallel, I've hooked up with a hardware guy in Europe to design and build (as Open Source) an accelerometer probe based on the same chip and an onboard MCU. The programmer for the MCU came in today so I can start porting my code. We also ordered 6 boards to prototype and test. Here's the basic design. The jumpers are to experiment with sensitivity and a couple of other options like printing in X, Y or Z so the board can be mounted in different orientations.
It's going to be a tiny little thing.

Impressive!

If the accellerometer is onboard, then this the universal version of the product after all? Or is it specific to the duet in some way?

That is looking great. You have done some excellent work, and I can't wait to see how it works.

Looks good.
What size are mounting holes? 4.2 mm dia. would be good in my opinion. (to fit onto stand off screw)
Also, the older SeeMeCNC aluminum stand offs where .375" dia., so you would want to make sure nothing shorts out with those.

Hey there. Hardware guy here. The holes are M3 at the moment (3.2mm hole in a 6mm clearance circle) but mounting is definitely a thing that will change with the final version.

The plan is to have the board act as a normal ends top, so it'll be fairly universal.

^^ What Duckle said!

I have most of the code ported over to the new MCU platform eval board. Next up is to get it to actually talk to the accelerometer. I ended up ordering the accelerometer eval board so I can work with it, these chips are teeny tiny.

Neat. Whats the ETA on getting the the first prototypes for in-house testing? If i'm understanding correctly this will just attach to the current accelerometer board?

Nope. We are talking about two completely things that I am working on.

1) a simple interface to the existing HE280 PCB accelerometer so the HE280 can be used with other (standard) firmwares
2) a completely brand spanking new accelerometer probe that does NOT require the HE280

Some more info:

1) I have a hand wrapped prototype ready to test hopefully today. I've already updated its firmware to take into account the new sensitivity setting in the new firmware. This device simply plugs into the 3 pins currently occupied by the I2C and interrupt on the HE28 PCB. At the other end, the 3 wires look like a standard 3 wire endstop. After I validate this works I am going to fabricate a little PCB to simplify construction. I will then offer it as a simple kit or finished product. The only parts are two 3 pin 90° edge connectors, an LED and the ATTiny85 preprogrammed with my code. It will be an Open Source project too.

2) This device is shown in the renderings above. It uses an accelerometer from the same family that the HE280 accelerometer is from. It uses a completely different processor that is ideal for applications like this. The boards are at the prototype house and should arrive here at the end of november. I have all of the components to assemble 6 prototypes to test. These are all spoken for. Once we validate the feature set and functionality we will make and design changes and manufacture these. It will be an Open Source project. I am working with Duckle (he's doing the hardware design) on this project. He's in Europe. The way we plan to market these is I'll handle all sales in the Americas and he'll handle Europe and the UK. I suppose either of us would handle PacRim and Australia - it's more a question of favorable shipping rates than anything else.

Hmm. Interesting. I was thinking you'd just add a few jumpers so you could either get the data from the on board accelerometer or the he280. But, i guess the second board requires a lot of additional components. I'm definitely interested in the first board.

Hmm. Interesting. I was thinking you'd just add a few jumpers so you could either get the data from the on board accelerometer or the he280. But, i guess the second board requires a lot of additional components. I'm definitely interested in the first board.

It is much much simpler to have a single purpose solution for the HE280. Literally everything can be done with 3 pins in and 3 pins out and an ATTiny85 in between. And it will slip right into the connector on the HE280 so there are no mounting issues. The full blown board with accelerometer would need to be mounted somewhere, wires from the HE280 to it and from it back to the whip. Much too complicated. And if I used the different processor, it would require many more components. And why incur the extra cost of that plus an extra accelerometer? It will be much easier to make and support as 2 separate devices too.

Yes, i didnt think about all the extra components the accelerometer needs.

When you have some of the HE280 Adapters ready to sell, I am excited to buy.

You haven't heard the price yet! That might curb your enthusiasm!

Tree fiddy?

g.

At least, maybe more

On the "serious news" front, the PCB fab just mailed the 6 prototype PCBs shown above for the free-standing accelerometer probe (no HE280 needed). I also have all the components to assemble them. I am working on the software and have the MCU eval board connected to the accelerometer prototyping board and am working on the I2C communications. Once that is done, the rest of the programming should be straight forward.

I wouldn't mind getting one of those - Orange Menace is awaiting her update and I would like to stick with my J-Heads on that machine...

g.

The 6 prototype PCBs came in. These are for a dedicated accelerometer probe, no HE280 required. I've made some progress with the firmware but learning a new SDK, IDE and processor eval board is time consuming. But I am talking to the accelerometer now. I just need to sort out how to enable and handle interrupts and then I should be good.

Very nice! Did you have those made at Osh Park?

g.

Yes, OSH Park. Quick and cheap.

and great quality... i tried one of the chinese guys for a large board i was making (arduino due shield) and the quality difference was noticible.

your planning to hand-solder those? I can't seem to master smd solding for anything smaller than 1208... the solder seems to go everywhere :-/.

I suspect there's going to be a toaster oven involved.

g.