Re: MCP CANADAPTER 12V? | Posts 16+

According to Robotnetz, either the ULN or both the ULN and my AVR might fail. Apparently, the ULN isn't supposed to be used for that purpose?!

Were you also told why?

The only statement I can make is that it's a driver and not a level shifter.
<!-- m --><a class='postlink' href='http://www.roboternetz.de/phpBB2/viewtopic.php?t=33738'>http://www.roboternetz.de/phpBB2/viewtopic.php?t=33738</a><!-- m -->

Okay. The ULN2003 is designed for TTL voltage levels (including at the input).
Der ULN2004 ist der CMOS Pendant zum ULN2003 und verträgt 6-15V.
ULNs essentially operate as open-collector circuits.
<!-- m --><a class='postlink' href='http://www.st.com/stonline/books/pdf/docs/5279.pdf'>http://www.st.com/stonline/books/pdf/docs/5279.pdf</a><!-- m --> On page 3, you can see the circuit diagram.
In my opinion, this should solve your problem.
The suggestion to use optocouplers (PC 847 for 4 channels, PC 827 for 2 channels, PC 817 for 1 channel) is also worth considering.
I had a similar issue with my control system back then.

The switching transistors must be fast types anyway.
You can also play it safe and use optocouplers there.
You just have to see how fast they are when switching.
There are probably some that are very fast.

I don't have time right now to look for specific ones.

Best regards, MaxxThe suggestion of using optocouplers (PC 847 for 4 channels, PC 827 for 2 channels, PC 817 for 1 channel) is still worth considering.
I had something similar in my control system back then.

The switching transistors need to be fast types anyway.
You can also play it safe and use optocouplers there.
You just have to see how fast they are when switching.
There are probably some that are very fast available.

I don't have time right now to look for them.

Best regards, Ma'I had a similar issue with my control circuit back then.

The switching transistors definitely need to be fast types.
You can also play it safe and use optocouplers there.
It just depends on how fast they are when switching.
There are probably some that are very fast available.

I don't have time right now to look for any.

Best regards, Ma'You can also play it safe and use optocouplers there.
You just have to check how fast they are when switching.
There might be some that are very fast.

I don't have time right now to look for any.

Best regards, MaxSure, Maxx.

Okay, I've now assembled my circuit, but it's not working as described . When I apply 12V to the input pin, I unfortunately only get 0.5V at the output .

Hast du die internen Pullup auf den Eingängen des µC aktiviert?
DaThe signal will definitely be inverted and connected to the input.
The ULN is an open-collector Darlington array.SGM

Thank you, the keyword was 'inverted' .

Okay, I still have one problem: as soon as I set the fuse bits for 'clocksel' to 1111 and 'su' to 11 on the ATmega microcontrollers, nothing works anymore . Normally, this should enable the external quartz crystal, but unfortunately, it doesn't start oscillating. The pin of XTAL1 is at a high level, and XTAL2 seems to be open. I then tried to revive the device by connecting an external clock source to XTAL1, but that didn't work . Now I've already ruined 4 devices. :(

If I use the internal RC, I can already play with CAN .

Bei den Fusebits kann ich dir leider nicht helfen, da ich mich damit noch nicht wirklich befasst habe.

It's already great that the print function works with the internal RC .

SGM

ok, war kein problem mit den fuses sondern ein layout bzw ein problem beim platinenfertigen, da war leider eine durchkontaktierung zu nahe an einer vcc leitung.
das problem konnte ich mit einen cuttermesser jetzt lösen.
I've already found suitable enclosures for my PCBs.

@maxx
With the AT90CAN128, I've optimized the drivers to the point where I can send messages in an idle state with a 0.9 ms clock cycle. Receiving, unfortunately, only works at 9 ms so far, and when receiving data, simultaneous sending is limited to 3 ms.

And with the board I have here (with the MCO2515), sending is also relatively fast, and receiving is around 15 ms. However, I've only written the program in BASIC so far and haven't optimized it for timing, which wasn't necessary for the purposes I had in mind.

How are things going with your AT90CAN128? Has your test board arrived yet??@maxx
With the AT90CAN128, I've optimized the drivers to the point where I can send messages with a 0.9 ms clock rate when idle. Receiving, however, is only working at 9 ms so far, and simultaneous sending and receiving when data is being received is limited to 3 ms.

With the board I have (using the MCO2515), sending is also relatively fast, and receiving is around 15 ms. However, I've only written the program in BASIC so far and haven't optimized it for timing, which wasn't necessary for the purposes I had in mind.

How are things going with your AT90CAN128? Did your test board arrive yet?How are you doing with your AT90CAN128s? Did you receive the test board yet?How are you doing with your AT90CAN128s? Did your test board arrive yet?

Okay, the issue wasn't with the fuses themselves, but rather a layout or manufacturing problem on the PCB; unfortunately, there was a via that was too close to a VCC line.
das problem konnte ich mit einen cuttermesser jetzt lösen.

That must have been incredibly difficult to find.
You have to be pretty creative to come up with something like that.


beim at90can128 hab hier die treiber soweit optimiert, das ich im leerlauf nachrichten im 0,9 ms tackt senden kann, empfangen geht bis jetzt leider nur im 9ms tackt und wenn ich daten empfange geht das gleichzeitige senden leider nur mit 3ms.

That sounds really great.
Did you write the driver in C?


und mit der platine, die ich hier hab(also mit mco2515) geht senden auch relativ schnell und empfangen so im 15 ms tackt, das programm hab ich aber bis jetzt nur in basic und noch kein bisschen auf timing ausbereited, brauchte ich aber auch nicht für die zwecke, für die ich das geplant habe.

I think it's great that this works. I was actually planning to build a similar low-cost circuit myself.

wie gehts dir mit deinen at90can128? dein testboard schon gekommen?

My test board arrived quite some time ago.
However, the renovation of our apartment got in the way.
I didn't feel like getting into the car that evening either.
Now I will have more time again and will dedicate myself to this topic.
I have another Climatronic control panel at home that I can use for testing.
Do you think I could make it so that the control panel works independently with my circuit?
I'm not experiencing the same problem as you are with your PDC control unit.

The problem in the automotive sector is that the input voltage (12 volts) can fluctuate... starting the engine, engine running, engine off, battery fully charged, battery nearly empty.
Da kann die Spannung bis zu 13 - 14 Volt betragen, aber auch etwas unter 12 Volt abfallen.
With the circuit attached, you stabilize the incoming voltage to a steady 5 volts at the output, which remains constant. This makes it well-suited for use with microcontrollers in a vehicle.

Greetings.

Benjamin.