AMB linestage rebuild

[Nick]

Yep, I saw that in the sheet, but not sure how long that pulse lasts as its only for the duration of the write signal, but if that was the case then why won't it drive your two inputs? Anyway, you know how to fix it so its just tilting at windmills.

[Thermionic Idler]

Yep, I saw that in the sheet, but not sure how long that pulse lasts as its only for the duration of the write signal, but if that was the case then why won't it drive your two inputs? Anyway, you know how to fix it so its just tilting at windmills.

Indeed, it's all finalised now - the parts order went off to Digikey this afternoon. This also gives me the option to extend the pulse duration a smidge if I want to - I just need to locate where it's defined in the software.

[Thermionic Idler]

New PCB populated and tested this weekend - all looks good! We now have a functioning input selection section and more optimised pull-up / current limiting resistor values.

Next weekend's task is to make up the ribbon cables to connect to the signal board relay coils (not quite sure why I didn't pick board connectors that interfaced to off-the-shelf ribbon cables, but hey ho, another design decision I have to live with...), test the signal board power supplies, then test the signal board itself.

Unless there's something silly like a fried op-amp, I have greater confidence that this will test out ok. I'm over the line with the complicated bit.

Next update in a week or so.

[Thermionic Idler]

Happy days - finally interfaced the relay coil connectors on the signal board to the driver board, the resulting relay clicks (rather than silence) was a good sign.

I then installed the two power transformers for the signal board left and right channel power supplies, both power supply boards tested perfectly. Then hooked them up to the signal board and did a signal test with the Analog Discovery.

And all looks good there - the volume control raises and lowers the signal level perfectly, the input selection works and the gain is identical in both channels. The frequency response of this thing seems to be flat to over 1MHz, with the 20Khz square wave response pretty much perfect.

The picture below was taken just before I completed the tests. You can see the signal board at the back.

This is good - all the key electronics have now tested OK, so it ought to be plain sailing from now on. Next weekend - install the XLR sockets and solder them to the board.

[Thermionic Idler]

OK, it's done. That was quite possibly the hardest thing I've ever designed.

Interior picture before the lid went on:



Front panel (getting the blackout film stuck on behind the perspex was a lark):



And the rear panel:



Now to get the system set up again and see if it actually sounds any good...

[Cressy Snr]

Well Dave, obviously, as you are aware, it takes a great deal of blood, sweat and tears to produce work of such incredible shoddiness.

I wouldn’t know where to start with that stuff.

[Thermionic Idler]

Well Dave, obviously, as you are aware, it takes a great deal of blood, sweat and tears to produce work of such incredible shoddiness.

I wouldn’t know where to start with that stuff.

Thanks Steve - over a year's worth of blood, sweat and tears in fact, looks like I started this thread last January.

Worth it though - it sounds - well basically I can't hear it, which is exactly what I want from a linestage. That slight midrange 'glare' that I had before has gone completely. The circuit itself is a very sound design (Bruno Putzeys knows his stuff), but I think some of it has to do with the fact that the whole system is now interconnected to the AES-48 standard, and is managing away RFI nasties very efficiently as a result.

[IslandPink]

Top notch !

[Mike H]

Excellent.