I had been waiting for some right angle IDC connectors and a few other things to come in the mail for ages now and eventually got sick of waiting and used some parts I already had in the workshop. So all this very excitingly this means that the mixer is now working (and sounding great)!
On testing, I decided to add a number of features for the next revision that will make testing and troubleshooting of the cards a much easier process. All PCBs now have a 2 pin molex connector footprint for applying power to individual cards during testing, as this can be a pain to do via the IDC connector. For now I have made a little jig on veroboard for doing this but it isn’t as elegant as just plugging in a connector and turning a power supply on. All cards will also have a jumper footprint for shorting the power and audio grounds together. I always design pcbs so that these two points are tied together at a very specific point (in this case it happens on the cue card). This means that when testing a single card I had to solder a resistor leg to join these together. Each of the input cards now also have 3 pin molex connector footprint for easy monitoring of the audio output. I would hope that none of these features are ever actually needed as they are only really useful during troubleshooting, but nevertheless they are now on the boards. The audio output on the input cards could allow for additional features such as level monitoring or direct outputs from the channels though these aren’t things I have plans for at the moment.
There are also a few things I have been agonising over lately: The line card output BJTs and the JFET used in the tone and summing amplifiers.
The output devices used in the line cards of the original mixer were a 2N6291 and 2N6108 complimentary pair, a fairly heavy duty pair. Other clones of this mixer have used the TIP31C and TIP32C pair for this task as the originals are difficult to obtain. I’m using the smaller BD139/140 pair in my version, though each are still capable of supplying up to 1.5A – more than enough for this application. I haven’t tried the TIP pair yet, but will do so once I order some in. The main differences between the pairs are the slower speed of the TIP series and the lower current capacity of the BD series.
I got my hands on a bunch of the original JFET (2N3822) used in the original so I have been using those in my remake, but I think I need to look at a potential replacement for others as they aren’t the easiest to come across. Considering these devices are just supplying a constant current for the BJT, they aren’t actually all that critical and I really do want to make this build as easy as possible for people. You can actually still buy these brand new from mouser at a fairly extreme cost, $44.32AUD each, so all 12 for $532.08 – not a realistic option at all. I purchased mine from an ebay seller based in the Netherlands for just under $3 each. I also notice a seller in Russia selling KP303 models for a slightly cheaper price but I haven’t had a chance to test these for myself.
I still need to order the revised PCBs and build another prototype up before selling anything but will keep everyone posted as to when that happens.