Building The Appendage. Share your Experiences

AlanP · Posted: Thu Mar 01, 2018 2:47 pm Post subject:

Is there a preferred vendor to order the Softpot from? (I'm thinking 500mm.)

I couldn't give a reference, but I vaguely remember an account of one vendor shipping a softpot after folding or creasing it to get it in the packaging.

elmegil · Posted: Thu Mar 01, 2018 2:50 pm Post subject:

I can only speak for mouser; it was packaged appropriately, but they didn't have the 500mm, only a 750 or the shorter ones.

Psynth · Joined: Jul 18, 2018 Posts: 19 Location: UK Audio files: 5

So, here’s the thing. I built the appendage on stripboard (don’t ask!), and it works completely – provided I warm the boards up with a hot air gun! The weird thing is, when it is cool (and depending on the ambient temperature), it works with the bottom couple of inches of the ribbon, but not higher up. As I warm the board up, the usable length of ribbon increases, until all functions work for the full length of the ribbon. Then, as it cools down, the top end ceases to function. (The usable length also increased slightly over time, presumably as current flowing in the circuit warms it up.)
I have traced the problem to the SLSHClk put out by AND gate U4 pin 3 (all references are to the 1.07J schematic, attached). The inputs to this gate, on pins 1 and 2 are the GatedClk, and, effectively, Comp – which goes high when the ribbon is pressed. I measure the GatedClk as a pulse with a frequency of about 1.1 kHz (my main clock runs at 11kHz), with the positive going pulse being 37us wide. That is present on pin 1 regardless of where, or whether, the ribbon is pressed, or the temperature. Likewise, pressing the ribbon, at any temperature and anywhere, makes Pin 2 go high.
The output on Pin 3 SLSHClk, ought therefore, to be identical to the GatedClk, when the ribbon is pressed. Which it is – at the bottom end of the ribbon or when I warm the board. But at the higher end of the ribbon, pin 3 remains Low. In a transitional length of ribbon, pin 3 goes High with a 1.1kHz pulse, but the width of the pulse decreases over half an inch or so of ribbon from 37us down to less than 1us and then to nothing.
Since the output on U4 pin 3 also ultimately drives all the S&Hs, and the sample deinhibit, it makes sense that the absence of a SLSHClk would indeed mean no sample.
Been tearing my hair out over this one. Any help welcome!
I have swapped the 4081 in U4, to no avail. I am using ST HCF4081’s – not sure if using CD4081 would help? My other thought was that the SLSHClk signal is fed into pin 8 of U8 (LF398) via a voltage divider consisting of 5k6 and a 1k8 resistors. That means there is a 7k4 path to Ground, which assuming SLSHClk is at 12v (when High), would seem to mean U4 pin 3 has to source about 1.6mA. The datasheet for the HCF4081 says the maximum output drive current at 10v supply is typically 2.6mA, but could be as low as 1.1mA. At 15v supply, the minimum is 3.4mA and typical 6.8mA. (I’m not sure if these are per gate or total for the whole IC). So I am just wondering if my IC, at 12v supply, is just on the cusp of being able to supply the necessary current – and presumably if it is unable to do so, the gate won’t function properly. Quite why, or how, this depends on both temperature and the voltage sampled by the ribbon (which as far as I can see is not directly in the various clocking signal paths) is a mystery.
Is there any reason why the voltage divider on U8 pin 8 uses such low values? The LF398 doesn’t need any significant input current to that pin, so why not use, say 56k and 18k? That would reduce the load on the AND gate.
But maybe there is some other explanation?

Thanks
Peter
(previously known here as prgdeltablues, but that account has stopped working for some reason)

appendage_pcb_schematic_rev107J_page 3.bmp

Description:

Download

Filename:

appendage_pcb_schematic_rev107J_page 3.bmp

Filesize:

1.77 MB

Downloaded:

424 Time(s)

elmegil · Posted: Wed Sep 26, 2018 8:27 am Post subject:

The first thing that occurs when you talk about heating up the board is solder joints. Heat makes things expand and could complete connections that are otherwise dodgy. Those are often hard to find by poking and probing because usually putting your meter probe onto something can *also* complete a dodgy connection.....

State Machine · Posted: Wed Nov 14, 2018 9:33 am Post subject:

Psynth · Joined: Jul 18, 2018 Posts: 19 Location: UK Audio files: 5

Hi,
thanks for the response. Been moving house, so not been able to work on synth recently. I've sort of parked this problem for a while - the board is a fiendish triple stacked stripboard with a rats nest of wires to the panel - not easy to get at components. But I think I can disconnect the output of the 4081, which should allow me to test the 'load' theory.

Peter

State Machine · Posted: Sun Nov 18, 2018 3:25 pm Post subject:

OK Peter, I'll be watching the posts. Get settled in and I guess we'll hear from ya! Smile

Standing by
Bill

AlanP · Posted: Sat Dec 15, 2018 5:32 pm Post subject:

Posted Image, might have been reduced in size. Click Image to view fullscreen.

Still need to order a few components, but almost done Smile

One of the pots for the front panel had a shorter shaft than the others, so I put it on the input attenuator at the bottom.

elmegil · Posted: Sat Dec 15, 2018 10:45 pm Post subject:

elmegil · Posted: Fri Apr 12, 2019 11:54 pm Post subject:

Ok.....

I got the panel PCB today.

It's built, hooked up, the appendage is attached via a chunk of stripboard & a jack, the softpot and the stripboard anchored on a chunk of oak....

I'm at Step 7 of the calibration, the external input.

I've triple checked that I have it set up as it says, and I am at

"Connect the positive lead of the DMM to TP1. Set the DMM to the very lowest range, and, while applying pressure to any point on the ribbon, measure the voltage present at TP1. Record that voltage here:

(note: this will be an extremely low voltage, in the low millivolts if it is present at all. Be sure to record the sign (if it is a positive or negative voltage)."

Except it's not

It's exactly the same voltage I would be seeing in "normal" mode. If I hit about the middle of the ribbon, it's about 5V.

I really don't grok this mode anyway. I'm trying to figure out what I'm doing wrong.

The other settings not mentioned in the setup are: ribbon mode is voltage, not current. Trigger mode is single. Input select is "ribbon".... hm.....

Switched it to sample and now it's "extremely low", -8.4mV. So perhaps the instructions omit that setting?

State Machine · Posted: Wed May 15, 2019 8:56 am Post subject:

elmegil · Posted: Wed May 15, 2019 9:34 am Post subject:

I did muddle my way through it and have used the appendage very happily since. However I don't recall what I had wrong either :/

Psynth · Joined: Jul 18, 2018 Posts: 19 Location: UK Audio files: 5

Update on my post of a few months ago. Finally got my synth up in the new house. A couple of modules didn't work - had to swap out two chips (why being moved caused them not to work??). And the appendage.... now appears to work just fine. The problem I reported has just gone away. Absolutely no idea why, the only thing I can think of it is now inside the house rather than in a shed. Was warm enough last September, but perhaps damper, causing some leakage currents, maybe just enough to make the 4081 struggle....So perhaps not the heat from the hot air gun making it work, but the drying effect?

Anyway, it is now working!

Peter

State Machine · Posted: Thu Nov 28, 2019 10:12 am Post subject:

Glad the Ribbon controllers are working for you guys ! Have not been here in a while so decided to check. However I am still very active but just not as publicly. I do miss the interaction and will make efforts to be in the forum much more. Have fun guys !!

Bill