Right, so I'm back from holiday and have made a good start on the boards (enough for one channel built). Next task is to get enough of the case built that I can safely run it (I've no desire to loosely connect anything up that is at mains voltage). What surprised me was dealing with SMD components is a lot easier than I expected.
My plan is to build and fully understand only one channel first and get it working, then complete the second channel after that. I'll post some photos once a bit further along..
My plan is to build and fully understand only one channel first and get it working, then complete the second channel after that. I'll post some photos once a bit further along..
Right, SMD is much easier than most of us expected before we tried it for the first time
Keep it up
I've got a new version of control board with an onboard control transformer and onboard relays for a single power supply amplifier. After some silly assembly mistakes it's working perfectly. It uses I2C temperature sensing instead of the MJE340, for actual temperature readings.
The 21st Century protection system is almost ready to become interactive. A proper USB interface and a serial port for Bluetooth/Zigbee/whatever is next.
The 21st Century protection system is almost ready to become interactive. A proper USB interface and a serial port for Bluetooth/Zigbee/whatever is next.
Attachments
+1Looking forward to use them in my two monoblocks
Thanks for your effort to make diyamps safer then commercial amps.
Both locations are slow to affect the sensor detection.
Attaching a sot23 transistor to the thick middle lead out of a power Output device is quicker and reaches a temperature closer to Tj inside the device.
Have you checked the pbc traces distance for safety between 230V AC and small signal part?
BR, Toni
Creepage distance for 240VAC on a coated pc boards is 23 mil. The 12V feed to the relays would be too close on a bare board, but non issue coated.
There are different schools of thought as to where to sense temperature. You can get instant temperature sensing of a single device by placing a sensor right on a device, or you can get average sensing of all the devices by placing the sensor right on the heatsink and allowing for thermal resistance. Placing the sensor on the clamping spreader bar is only a good reading in dead still air, so not really useful at all (not really useful for bias sensing either). We aren't really interested in the instantaneous temperature of a single device. the average temperature of the whole output stage is more useful to us. With the parts values listed in the BOM, the amplifier will shut off when the heat sinks reach 65C.
Is this Canada standard? I think 23mil distance on pcb board between 230V AC and small signal is too less even if coated. The IPC2221 standard for 301 - 500 V is 800 µm (32 mil). I'm sure between 230VAC (can be up to 253V and 357V peak) and small signal you have to add additional safety distance. (For the EU safety rules I think there is additional distance necessary; unfortunately the documents are hard to find and cost extra money).
The board is big enough to add extra safety...
BR, Toni
Yes that was a typo. I meant to say 32 mil. The traces aren't anywhere near that close. If anyone is concerned about it, they can remove the top unused mount lead on the relay.
