PIC32 pin to pin replacement for arduino!

Just about fell off my chair when I saw this! [www.digilentinc.com]

PIC32 pin to pin replacement for arduino.... for $50! ($10~15 less than arduino mega)

They also have a PIC32 pin to pin replacement for the UNO.

There was another mention of this acouple of days ago

[forums.reprap.org]

And I know of one other Arduino-Compatible board, but I don't think they have a MEGA equivalent

[leaflabs.com]

I've ordered the chipKIT 32 to test out with Sprinter. We'll see how it goes.

I just received one to also test with sprinter. It has 3.3v I/O so now I'm not sure it will be easy to get it all working with the stepsticks and the ADC measurement will be out

I have not yet had time to power mine up but have downloaded the very large Dev software for arduino compatibility.

Also I am not sure if it will actually be any faster than a mega when in arduino compatibility mode. You may need to run the microchip compiler to get it working at full speed

I think Forrest Higgs is using one in his new Dev

---

[richrap.blogspot.com]

I am in the process of porting sprinter to the Max32. Its a little more complicated than just uploading the firmware to the board. There was a lot of code in Sprinter that is AVR specific and will not work on the chipKIT out of the box. Not even all of the arduino libs are working on the chipKIT at this point. I know that the Servo lib is one of them. I have a post about it in the RAMPS forum and we have been talking about it for the last week in the IRC.

---

Charles Gantt.
Build blog [themakersworkbench.com] <---- my personal site
Build worklong [www.thebestcasescenario.com]

Mine should be here in the morning.

The A4988 supports 3.3v logic supply so only only ADC needs attention

Actually, since the temperature measurement is done with the ratio of two resistors and we're using vcc as a reference, it doesn't actually matter what voltage Vcc is! The ADC readings should be fine.

The part that worries me is the mosfets- they won't be turned on fully at only 3.3v, so they'll dissipate too much heat and explode, especially the bed mosfet which is already struggling. They'll need some sort of driver. A max3232 would work, as would some sort of discrete buffer driving the gate from 12v similar to this schematic

---

-----------------------------------------------
Wooden Mendel
Teacup Firmware

Unless you need fast switching you can do the level changing with a single small signal MOSFET as I did here: [hydraraptor.blogspot.com]

---

[www.hydraraptor.blogspot.com]

I created a IRC Channel for chipKIT RepRap Dev. its #chipkit

---

Charles Gantt.
Build blog [themakersworkbench.com] <---- my personal site
Build worklong [www.thebestcasescenario.com]

I'm using logic level mosfets so they only need about 2.1~2.5V for the gate(IR Part IRFZ44N) pretty sure any logic level mosfet will work.

Alternatively you could go the route nophead is using, or if speed is a concern you could use a a general purpose npn like a 2n3904 to drive the gate of the mosfets.

IRFZ44N has a worst case gate threshold of 4V and RDSon is quoted at 10V.

---

[www.hydraraptor.blogspot.com]

Vthresh is where they just start to turn on. To carry any decent current, you need to find the Vgs where the Rds(on) or Ids flattens out. It's quite a bit higher than Vthresh!

Good examples of logic level mosfets are IRL3803 or IRLB8748 which can both carry about 15A at Vgs=4.5v. Even these aren't much good at Vgs=3v though!

---

-----------------------------------------------
Wooden Mendel
Teacup Firmware

I have always felt that mechanical relay switches are far Superior to silicon when it comes to turning on and off relays. also you know by the sound of latching that it is working. you also know the fail state of mechanical switches.

However I can't make a mechanical relay PWM at frequencies above my hearing range

---

-----------------------------------------------
Wooden Mendel
Teacup Firmware

Also if you calculate the lifetime switching high current every few seconds it will only be a few months.

They can also fail open or closed.

---

[www.hydraraptor.blogspot.com]

10 times a minute, [search.digikey.com]

tested for 10,000,000 ops 300 times a second.
MECHANICALLY 10,000,000 ops. at no load.


so for 1 year 10 times a minute 3110400 latches. tests prove over 2 years life constant use. (given a reasonable use of printer 36hrs a week, i would give the relay an average life of over 14years!!!! in theory though. we know that the smoke god visits relay's as well.... also this is a test for latch contact, with no current flow. at max current it could be as little as 100,000 cycles, of course you would never put the max contact rating into a relay. 5 amps or less with this one. the coil will outlive about any mosfet, especially because of electromigration of the materials due to current flow, and high temp, unless you have heat sinks installed. [en.wikipedia.org]



. I've been using relays for near 2 years for heating coils, and powering very power hungry objects. the only time I've had to replace anything it was the transistor that powered the relays, and both times it was from a power outage. building in some transistor protection ensures that the transistors are fine no matter what. never needed to replace any relay yet..... don't want to jinx myself..... knocking on wood...


again this is my own opinion. relays deserve a better place.

Unless you use a ridiculously over specified relay the life is going to be much closer to the electrical life than it is to the mechanical life. So say 500,000 for a heavily loaded relay at 10 per minute would give you about 35 days of life.

I don't think metal migration is an issue with high power MOSFETs, the metal parts are too massive. It is an issue with VLSI where the features are very small.

My Mendel has run 24/7 for ~17 months. It has lots of MOSFETs in it running at high temperatures and none have failed yet. The things that have worn out are plastic extruder parts and cables. A relay would have done 7.5 million operations, an order of magnitude more than the electrical life, so I would expect it to have failed.

---

[www.hydraraptor.blogspot.com]

In the mid 70's I was testing DIL reed relays to destruction they only ever managed between 250,000 to 300,000 before failure. They were running under load 300mA tested right accross their full temprature range (very cold to very hot) also their full air pressure range including a 2 torr vacuum.. A failure was indicated by a very high contact resistance/open circut or shorted contacts.

I don't think there has been any significant imporovement in relay contact technology to make them any better than they were back then so I also do not think relays are a reliable option paricularly with the currents we are looking at.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

> The part that worries me is the mosfets- they won't be turned on fully at only 3.3v, so they'll dissipate too much heat and explode, especially the bed mosfet which is already struggling. They'll need some sort of driver. A max3232 would work, as would some sort of discrete buffer driving the gate from 12v similar to this schematic

Would this driver work as a rework to RAMPS to allow it to use the PICKit MAX 32. Were any other issues discovered. I'm rather motivated to get it to work since I'm running out of steam on the Arduino mega I'm using for a project. Would the RAMPS still work with arduino mega if this modification? I'm an EE neophyte that can solder!