ES9018 I2C controller

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I decided to use the Atmega644 instead of the 328. It has double the programming space and 5 times large EEPROM then the 328. This should be enough for now. This chip is really large, but I will have no problems soldering it :D

I added an ISP header to the design to simplify programming the chip by using a programmer like Arduino, AVR dragon or similar. At first I wanted to have as many functions possible on the controller, but I'm now going for an approach with several add-on modules. The first two add-on modules would be the push buttons and rotary encoder boards. Later on I'll will get busy with ethernet module and maybe some more add-ons.

The I2C port expander has been removed from the design. The Atmega644 has lots more IOs compared to the 328, so it should be enough. Otherwise I can always make an port extender Add-on.

The controller has to be idiot proof (short protection, protection agains high voltages etc). The Arduino uses fi. a resettable fuse. This is nice, but I'm not sure which path to take here. Any ideas would be welcome.
 
Overvoltage and current protection

I've been looking at simple but effective solutions to make the controller idiot proof.

I think that a crowbar circuit will provide enough protection when the right Zener diode is used. It is fast, easy to build and uses cheap components. I do not know what the tripping voltage should become, ideal would be 5.7 Volt imo. According to the Atmega's datasheet its maximum operating voltage rating is 6V.
I'm thinking of using a circuit similar to this one: Crowbar circuit
The fuse in this circuit must be a resettable type, preferably an SMD version like the Bourns MF-SM Series or MF-NSMF Series. Perhaps the MF-NSMF Series is better because it has lower resistance and take up less space.
Datasheet for Bourns Resettable fuses


The safe operating area of the 644 goes from 4.5V to 5.5V. To protect it from under-voltage I'll use the chip's brown-out fuse/function and set it to 4.3V

In the first post I mentioned that I was not sure about adding a voltage regulator circuit, bit perhaps in the perspective of protecting the chip it is best to use it. In combination with the crowbar circuit it should provide adequate protection, but it would also mean that the circuit needs to be powered by 7.5V to (lets say) 12V instead of 5V. :trapper:
 
Why do you want to make the controller idiot proof?

This is DIY, if someone puts 12v on a 5v micro controller board he is a dumb fool and should be collecting stamps!

And why such a hassle with crowbars? Just put in a 7805 and feed it from a 12v supply. All the protection you need.
 
Why do you want to make the controller idiot proof?

This is DIY, if someone puts 12v on a 5v micro controller board he is a dumb fool and should be collecting stamps!
Well, that's your point of view. A mistake is easily made. Better safe than sorry.

And why such a hassle with crowbars? Just put in a 7805 and feed it from a 12v supply. All the protection you need.
I'm just coming up with some ideas so they can be discussed. Personally I like the extra safety precautions to protect the chip. The crowbar is a cheap but effective solution, but maybe there are some other interesting options too for adding that little bit of extra protection.
 
you might have a look at the fifo, it has some pretty excellent protection short circuit, overvoltage, reverse polarity etc. with SMD resettable fuses as part of it. it also makes use of SMD LED indication for all parts along the main signal paths that indicate correct connection, speed, lock from one stage to the next etc. so its really easy to track down a problem.

the modern dac, of the type this board is going into is a pretty complex beast, in fact I would say its easily the most complicated component, multiple supplies, many quite similar. sure its always best to check thrice before doing anything, but sometimes **** just happens and ds23man, if you arent making the odd mistake in DIY, you arent doing it right...
 
every idiot on the street corner is going to build an automation system for a 1.5-2k dac? Elitism is out of place here, its already a select group that will even attempt a project like this.

I dont know about you, but I learned connecting things the wrong way was not the best way forward some time ago. now that the novelty has worn off, I dont mind if it doesnt cost me money/time :D

I will continue to learn new things and sometimes make mistakes that teach me something new, or reinforce good working practice. With basic things like these protections, particularly in cases like this where there is no penalty for sound quality and minimal extra expense, I see it as good design. mistakes can and do happen
 
you might have a look at the fifo, it has some pretty excellent protection short circuit, overvoltage, reverse polarity etc. with SMD resettable fuses as part of it. it also makes use of SMD LED indication for all parts along the main signal paths that indicate correct connection, speed, lock from one stage to the next etc. so its really easy to track down a problem.
Thanks! I'll definitely have a look. I have seen the SMD LED indication before. The Hypec Ncore power amps have them too. It sure is a nice feature.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.