Missing negative voltage 3

Hi
I’m assembling my SMR-4 board and as few other people here, the power supply test failed at green points where the voltage was 0V instead of -5V (the other points are as should be). I have checked the polarities of parts several times and resoldered most of the joints. Reading the forum at other similar cases it seems like it has something to do with L7905 or LT 1054, or the caps around it. Before ordering new parts, I would like to be sure if and which parts might be broken.

The incoming voltage is 15,6V, all the pins on the LT 1054 are 0V except the 8th (the rightmost in the lower row). all the 3 pins of L7905 are 0V. At one point during the voltage testing there was some smoke somewhere near LT 1054, I disconnected the power cord. However, trying to reconnect the power again there is no smoke but the voltages are unchanged. The negative voltage problem was already before the smoke.

Could you help please?

There’s a very good chance you have simply destroyed the LT1054. The build instructions specify “Connect a DC supply (unregulated, any voltage between 7.5V and 12V, center pin positive)” and there’s a reason for that: the LT1054 does not accept voltages higher than 15V

It seems the text on my power adaptor is lying, it says 12 v but is actually 15,6V. It also (luckily) seems I hadn’t actually inserted the LT1054 in the socket since I thought the socket to be the converter itself (so much of my knowledge of electronics). However, this way I didn’t distroy the actual converter with 15,6V power, so now the problem is solved - connected another adaptor with 12V and negative voltage works fine.

Have you measured the first power adaptor under load, ie connected to the Shruthi-1 or anything that consumes amps? It could still be perfectably useable power adaptor… on the other hand if it shows 15,6V under load it’s ready for the /dev/null…

Weirdly I was reading the power supply chapter of Nicolas Collins fantastic ‘Handmade Electronic Music’ today during my lunch and one of his rules was never trust the writing on the front of a power supply, for this exact reason. He was recommending testing all of these power supplies before use because this is a common issue. Weird cause I’ve never even thought of this being a possible problem, and then this comes up on here! Glad you got it sorted out.

I have never observed more than a 3V difference - that’s why I recommend 7.5V-12V knowing that the LT1054 doesn’t take more than 15V. Maybe I should recommend 7.5V-9V…

i also have one that’s specified 12V and outputs between 15 and 16. it has a lot current output though, not sure if this may be a reason for the higher V

You’ve got to try it when loaded. There’s no other solution with a linear, unregulated power supply.
Given that you can expect a shruthi to require ~200mA under 9V, you can simulate it by using a 47 or 68 ohms resistor. Beware, that the resistor has to handle at least 3W.
Once you’ve loaded your linear power supply with this 47 ohms resistor, you can really know what good is your power supply for a shruthi.

I had a 7.5V, 300mA linear power supply which looked perfect for my shruthi. Unfortunately, it wasn’t beefy enough and the 7805 would not do its job properly. Still I can use it, but I’ll have to replace the shitty diode bridge (1N400x here) by a better one (using “beefy” schottky diodes).

Might seem silly, but make sure you have the 7905 the right way around. If you use the TO-92 part (the small one), the datasheet can be a little confusing because the image shows it from the BOTTOM! So ironic that this happened to me today, while building a 2164-based dual VCA.

I have a similar problem, which only turned up after I had assembled the filter (SMR4 mkII) and control (XT) boards.
For awhile, the control board was fully powered (display, menus, all pots & buttons working), but I never got an audio signal out. Then, today, I was going back over the filter board to check the audio signal path and all of the sudden the control board gets no power (except for a brief flash of some random pixels on the display, and a few times all LEDs lit up and stayed on…).
When I tested the power supply I got the same readings described in the original post, no negative voltages.