I'm more interested in this than the amps themselves. 🙄
You don't know the massive "22 century" leap you would make with the thermal/
overload monitored by using the analog. Rails ... too .
Here , you could script the "trip" points .... allowing for individual amp customizations.
The Solar power people not only use these analog inputs , but tie a SD card
to the ATmega to log long term analog fluctuations (with time stamp).
OS
Yes, using analog inputs for monitoring / measuring certain parameters, will be the next "leap". Built-in ADC allows doing it quickly and afficiently, enhancilg capabilities to infinity 🙂 Imagination is the limit 🙂
Thanks to Jeff for " tearing Arduino to pieces", so now we can use ATMega chip separately, making the board smaller and cheaper
P.S. I have also recently developed a solution, utilizing 74HC595N shift registers, allowing increasing the number of digital outputs to as many as required (8 additional outs per register).
Cheers,
Valery
Thanks to Jeff for " tearing Arduino to pieces", so now we can use ATMega chip separately, making the board smaller and cheaper
P.S. I have also recently developed a solution, utilizing 74HC595N shift registers, allowing increasing the number of digital outputs to as many as required (8 additional outs per register).
Cheers,
Valery
Is it a good idea to have one microcontroler do all that or would it be safer to have one do all the analogue processing and use another as a dedicated shutdown brain?
I thought about it as well. As far as the firmware is reasonably simple, it's ok to have it doing everything. But if we go for some relatively continuous analog tracking / processing, it's a good idea to have one chip mostly for startup/shutdown main cycle, and the other one doing all the rest continuous stuff independently, allowing protection to do its job very fast. ATMega chips are cheap 😉
I'm still kicking around the idea of datalogging. A Raspberry Pi might be a good option because of the ease of networking. I have no idea of coding requirements though. That's something I've never had the time to learn.
Do you think it would be safe to use analogue inputs to measure output device current? It would save quite a bit of external circuitry but the optoisolators are nice insurance. I would like to see thermal sensing done by the Atmega chip. It makes adjustment of parameters much simpler.
Ah, right - a good task for a separate board with network interface. Then, we can arrange a portal and collect the logs there. If something goes wrong, it will be easy to analyse the historical log and see the issues caused the problem 😎
Yes, I think some simple, relatively linear opto-decoupling circuit will be more than enough. It can also perform some "integration" (RC), so that you measure not the raw audio signal, but somewhat averaged values.
Do you know of an opto device that is capable of this task? My limited experience with optoisolators has always been digital. I've always assumed an AD conversion was necessary.
The Opto-Decoupling looks like way more trouble that it's worth. Extra op amps and an isolated supply.
For a very good linearity - yes, op-amps are required. But for our purpose - some simpler circuits may be tried...
A good example of a linear device - HCNR200 and HCNR201 (High-Linearity Analog Optocouplers). In the worst case, LM393 may be used for "normalizing" the receiving side of it (2 op-amps per chip).
A good example of a linear device - HCNR200 and HCNR201 (High-Linearity Analog Optocouplers). In the worst case, LM393 may be used for "normalizing" the receiving side of it (2 op-amps per chip).
The isolated supply required for an op-amp or comparator is the part I don't like. If it could be done with just an LED input it would be worth exploring. I'll look into those devices.
That's another option. It could become a little more involved with the supply being shared with multiple channels. Positive rails and negative rails would likely need to be monitored separately. Emitter resistors are a convenient source for sensing with no added hardware to the amplifier other than two wire connections.
Hello
greetings nice to see folks thinking of short circuit and overload protection
warm regards
Andrew
greetings nice to see folks thinking of short circuit and overload protection
warm regards
Andrew
Hi Andrew, in fact, overload and short-circuit protection for the amp is already implemented in existing design. We are now thinking about monitoring the rails and some other parameters in time with appropriate logging 😉
Jeff, the opto-couplers are linear enough, so even if we use op-amps for "normalization", it will only be required on receiving side. I mean - no isolated PSU required in any case.
Cheers,
Valery
Jeff, the opto-couplers are linear enough, so even if we use op-amps for "normalization", it will only be required on receiving side. I mean - no isolated PSU required in any case.
Cheers,
Valery
They look very cool! Such a tall PCB for them - is the idea to allow some heat dissipation?