Just fixed my Ibanez TS-808 pedal.

[DrNomis_44]

I just gave my TS-808 pedal another test out a few minutes ago and it is still footswitching reliably, so, I'll call it well and truly fixed now.

Who would have thought that one little 10nF ceramic disc capacitor would cause all that trouble?...but there you have it folks, funny thing is that the capacitor looks perfectly normal to me, that explains why the cause of the footswitching issue was difficult to find, faulty components don't always look like they're faulty.


Here's a pic of the faulty cap which caused the unreliable footswitching issue:




So how did I determine it was faulty?, I used my digital multimeter set to it's highest resistance range (2000k, or 2M) to measure the DC resistance of the cap, which in theory should be infinite, when I measured it I got a reading of about 1.8M, I compared that reading to the reading I got when I measured the DC resistance of the yellow cap (a known good one) that I used as a replacement, and got an infinite resistance reading, I then soldered-in the replacement cap.

In the schematic I posted, there's a 68k resistor going from +V,this resistor is connected in series with a 22k resistor which connects to circuit ground, using my digital multimeter, I measured the DC voltage at the point where the 68k resistor and the 22k resistor connect to each other (it measured 2.28V), there's also a 1M resistor connected to that point, and this connects to one of the contacts of the mechanical footswitch via a 22 Ohm resistor, I found that the DC voltage on the 22 Ohm resistor side of the 1M resistor was measuring about .5v to about .8 v when the footswitching was not working properly, now what's really interesting is that when I removed the faulty cap, the voltage at that point of the circuit read about 1.6v, so, my conclusion was that something about the faulty cap was reducing the voltage when the cap was in-circuit, I can now understand what was going on, the 1.8M DC resistance of the cap in series with the 1M resistor was forming what's called a "Voltage Divider" in electronics, normally the DC resistance of the cap should have measured infinite, but, it was actually measuring 1.8M, the 1M resistor is acting as a "Pull-Up" resistor for the mechanical footswitch, when you press down on the footswitch it causes the normally 1.6V on the 1M resistor to drop to 0V, this acts as a trigger-signal for the rest of the footswitching circuitry, because the normally 1.6V was being pulled-down to about .5v by the faulty cap, the rest of the footswitching circuit wasn't being triggered properly, and seemed to be stuck in one state, pressing down on the mechanical footswitch had no effect, as soon as I replaced the faulty cap with the new yellow cap, the voltage measured 1.6v again, and that restored the footswitching to normal operation.

[Simon Barden]

Have a look at the old cap and see if it's got any voltage value on it. Caps can go short or partially short-circuit if their voltage rating is exceeded. They might have used a cap with too low a voltage rating. That cap is there to take any voltage spike, caused by the foot switch, to ground. DC is a lot worse at causing switching spikes than AC and they normally occur on break of circuit when you get a very brief arc as the current continues to flow until the air gap gets too large to sustain it. The switching spike can be many times greater than the switched circuit voltage, so unless the replacement cap has a decent voltage rating, it's likely to occur again over time.

[DrNomis_44]

Have a look at the old cap and see if it's got any voltage value on it. Caps can go short or partially short-circuit if their voltage rating is exceeded. They might have used a cap with too low a voltage rating. That cap is there to take any voltage spike, caused by the foot switch, to ground. DC is a lot worse at causing switching spikes than AC and they normally occur on break of circuit when you get a very brief arc as the current continues to flow until the air gap gets too large to sustain it. The switching spike can be many times greater than the switched circuit voltage, so unless the replacement cap has a decent voltage rating, it's likely to occur again over time.

That's a very valid point, the pedal itself only runs on +9V DC, and I'm guessing that the voltage-rating of those small disc ceramic caps is about 50V DC or so, as far as I can make out with my eyes alone, there is no marking on the cap indicating it's voltage rating, other than the three-digit marking indicating it's capacitance value, the replacement cap has a DC voltage rating of 100V DC so I think that it should be more than enough, now that you mention it, I was using a Switchmode Power Supply to power the pedal at one point before I noticed the footswitching issue, the Switchmode Power supply was designed to put out 18V DC at 2A, but, it was plugged into a small box that regulated the 18V DC down to +9V DC, and the pedal was getting regulated +9V DC.

It's possible that the ceramic dielectric of that ceramic cap may have gotten punctured by a spike coming from the Switchmode Power supply, but then again if that actually did happen then that would have taken-out all the other caps and the JRC4558D Op Amp IC too, I'm thinking that somehow the faulty cap might have suffered from moisture ingress, since the weather has been a bit humid and wet lately, and the moisture ingress is responsible for the 1.8M resistance reading I got, well it's about the best theory I can come up with so far, as a side note, I noticed that the cap seemed to be a bit temperature-sensitive too, desoldering it from the circuit board seemed to temporarily make it come good...how's that for a weird one?

[Simon Barden]

You can get caps rated to just 2.5v or less. 100v should be enough though. The rating may be a code as it's a small cap. You can find the codes here: http://www.radio-electronics.com/inf...r-markings.php

[DrNomis_44]

You can get caps rated to just 2.5v or less. 100v should be enough though. The rating may be a code as it's a small cap. You can find the codes here: http://www.radio-electronics.com/inf...r-markings.php

Cheers mate, I'll see if I can look up the voltage rating of the cap later on this week, maybe tomorrow after I buy a magnifier from Jaycar Electronics, been meaning to do that for a couple of weeks, will let you know about my findings.

I just remembered that I have a 20Mhz Dual Trace Oscilloscope that features a component tester function, I could use it to test the cap and see what results I get, just need to make up a set of test leads for it first though, might post some screenshots tomorrow when I get a chance to do so.

[Simon Barden]

If it's got a capture facility, you could use the scope on the foot switching line to see if you do get occasional spikes and how big they are. Although having had a closer look at the schematic, I feel it fairly unlikely that they would be anything serious. Plus there are two other 1000pf capacitors connected to the flip-flop circuit that would see the same voltage spikes (if any) and should be affected in the same way (unless these had higher voltage ratings).

I'm more of a mind to think that it was just a single duff capacitor.

[DrNomis_44]

If it's got a capture facility, you could use the scope on the foot switching line to see if you do get occasional spikes and how big they are. Although having had a closer look at the schematic, I feel it fairly unlikely that they would be anything serious. Plus there are two other 1000pf capacitors connected to the flip-flop circuit that would see the same voltage spikes (if any) and should be affected in the same way (unless these had higher voltage ratings).

I'm more of a mind to think that it was just a single duff capacitor.

I'm in the same frame of mind as you are about the duff cap, unfortunately my Oscilloscope doesn't have a capture facility, it's one of those old school analog CRT types I was able to buy before more expensive digital storage types became available, but, I really want to buy myself a new Digital Scope some time this year, Jaycar Electronics stock a 20Mhz Digital Scope that retails for something like Au$500.00 or so, they also stock a 100Mhz Digital Scope, but it's just over Au$1000.00.

One other thing I did was to replace the stock 100uF/16V Battery Bypass Cap with a new 330uF/16V one, it's not going to have any detrimental effects on the behavior of the TS-808 Circuit, it'll most likely improve the 9V supply rail filtering.

[Simon Barden]

A 100MHz scope will tell you a lot more about the waveforms of short transients than a 20MHz scope will. Just a shame there's such a big price difference. You can get dual channel 100MHz (or even 200Mhz) scopes from China for nearer Au$750 (which I believe should make it duty-free for you) but I've got no idea how good they are.

[DrNomis_44]

A 100MHz scope will tell you a lot more about the waveforms of short transients than a 20MHz scope will. Just a shame there's such a big price difference. You can get dual channel 100MHz (or even 200Mhz) scopes from China for nearer Au$750 (which I believe should make it duty-free for you) but I've got no idea how good they are.

It's definitely worth looking into though, I wonder if I could get a good 100Mhz Digital Scope on eBay for a good price, I'll have a look a bit later on tonight.

I may even be able to buy the 100Mhz Digital Scope from Jaycar if they will allow me to pay it off on layby, I'll ask them about it when I pay them a visit tomorrow.

[Simon Barden]

It's all going to be about how many extra features they have, plus sampling rates. A lot of them seem to have very high sampling rates (1Ghz+) so that their anti-aliasing filter can have a gentle slope, so not ring and affect the higher frequency images.