Issue with late 70's Boss CE-1, sounds like parasitic oscillation. ???

[Gtr0]

Hello Folks.
I have this late 70's original Boss CE-1 that I have had lying about for years. I finally decided, because the chorus does sound amazing, that I would start using it.

I have tried this through two different amps and I am getting some strange behavior. In a 50watt plexi style amp it sounds like parasitic oscillation on the edge of the notes... this is the clip labeled "Amp-1". The "Amp-2" clip is a more modern-ish three stage + cathode follower amp - through this amp it just sounds like uncontrollable feedback.

I realize this is a long shot - but I have no idea where to even start troubleshooting. The chain is Strat (lace pickups) to Boss to amp... nothing else in between. I'm using a Reactive Load/IR system, so no chance of actual feedback. But even plugging the Boss in line, when plugging in to the output of the Boss (when the amp is on) just touching the tip of the cable to the jack, it goes nuts... again sounds like oscillation.

Any ideas are welcome - I figured it never hurts to ask, that maybe someone experienced this before...

I have attached two mp3's and an image - as far as I can tell, the unit is all original. Many thanks for even looking!

[pdf64]

If you’ve got an ESR meter, check all the ecaps decoupling the power supplies.
If not, tack in an equivalent cap across each in turn until a culprit is found. Or maybe just shotgun replace them all.

[Gtr0]

Thanks, PDF64. I figured that replacing the electrolytics would be something I'd have to do eventually.... I suppose no time like the present! :-)

Thanks!

[Gtr0]

I finally received all of the caps to re-cap this... but out of curiosity... something I have always wanted to know is what is an acceptable ESR reading? Mine measures capacity and series resistance of course, but how much resistance is considered to be too much?

[pdf64]

I think the point at which increasing the ESR causes an actual problem depends on the particular application, so there’s probably no hard and fast guideline for differentiating good from bad.

[Mick Bailey]

There are tables with approximate typical ESR readings for different cap types and values, but any cap with a significantly high reading is likely to be in decline and will worsen over time. Because electrolytic caps are used in relatively non-critical areas there's a wide latitude for ESR. In DC blocking applications the resistance can rise quite a lot before it becomes a noticeable problem and in the case of an amplifier stage coupling cap will result in a gradual loss of volume over time. If the cap is followed by a resistor to ground then the ESR will act as a voltage divider and the signal attenuation be more pronounced. I now see many early Boss pedals where the output is very much reduced and some of the smaller caps read into hundreds or thousands of Ohms. If a cap is used for AC decoupling then there can be stability or noise issues when the ESR gets above a critical level. If the cap is used for AC bypass in an amplifier stage then the gain will gradually reduce.

With ESR is the change can be very gradual and not noticeable until there's a very marked failure with the equipment and then you find that almost every electrolytic is on its way out. Bear in mind that ESR values for a new cap are typical and vary between brands (ignoring low ESR types). An issue is that when you start looking at ESR you frequently find that in a working piece of equipment you may find a lot of the caps are way higher than they should be. The question is then do you go ahead and replace them all, or wait until you have a definite problem? I see rising ESR as a life limiting condition for the cap,and an early indication of eventual failure, but this can take quite a while to manifest itself. I had a run a few years ago with Takemine preamps with no output at all where the SMD coupling caps were practically open-circuit. When I had working guitars in, I did some comparative checks and the caps in those were heading the same way - all eventually doomed.

[Vertigorocks]

Here's a Picture off my old trusty Dick Smith ESR Meter I built as a Kit for $40 Bucks back in the early 90's!! One of the best pieces of gear I ever owned!! Use it more than my Sencore LC102!!

[Helmholtz]

Ecaps are often used to provide a low impedance path (e.g. to ground) for AC currents.
In those cases you want the ESR, which acts in series with the capacitance, to be lower than the capacitive impedance at the highest frequency present.
Capacitive impedance is given by Zc = 1/(2*pi*f*C).
E.g., C = 10µ, f = 10kHz => Zc = 1.6 Ohm.

https://www.allaboutcircuits.com/too...ce-calculator/

[Mick Bailey]

Were Dick Smith kits pretty rare outside Australia back in the 90s? I have a ring tester built to a Dick Smith design that has a slight change to work with tube amp transformers. I really liked the practical designs that were (are?) offered and how well thought out they were. Sadly, I never saw any of the kits sold here in the UK and shipping from Australia was out of the question.

[Gtr0]

Ecaps are often used to provide a low impedance path (e.g. to ground) for AC currents.
In those cases you want the ESR, which acts in series with the capacitance, to be lower than the capacitive impedance at the highest frequency present.
Capacitive impedance is given by Zc = 1/(2*pi*f*C).
E.g., C = 10µ, f = 10kHz => Zc = 1.6 Ohm.

https://www.allaboutcircuits.com/too...ce-calculator/

Excellent thanks.

And thanks to all for the great info.

Unfortunately for me, I replaced all the electrolytic with exact values of Elna caps and the noise persists. Plugged into an Orange Rockerverb, the sound is the same as "amp 1" attached in the op.

One thing I did notice is that the Hi/Lo switch stopped working... in "Hi" it produces no sound. In low, it is as stated. I never noticed this before and have had it in "Lo" for a few years... I see switches on these units are frequently replaced.... but I also assume that would not cause the artifacts that I am hearing.

It seems perfectly fine when on the clean channel of the amp, until I crank the input knob on the CE-1. I hope it is none of the ICs or transistors...

Thanks.

[Helmholtz]

I'm using a Reactive Load/IR system, so no chance of actual feedback.

That only eliminates acoustic feedback.
You might have electrical feedback by radiation (caused by unsufficient shielding)

Always mount the bottom plate of the chorus for tests. Also try changing distances between chorus, amp, dummy load, cables and guitar.

Can you scope the oscillation?

The high setting adds a transistor gain stage at the input having low input impedance.
With guitar it's typically used in low setting.
I replaced the 50k input gain pot with a 500k pot to increase the input impedance.

Hard to tell if the broken sensitivity switch is part of the problem.
Measure it to find out what's wrong.

[Gtr0]

That only eliminates acoustic feedback.
You might have electrical feedback by radiation (caused by unsufficient shielding)

Always mount the bottom plate of the chorus for tests. Also try changing distances between chorus, amp, dummy load, cables and guitar.

Can you scope the oscillation?

The high setting adds a transistor gain stage at the input having low input impedance.
With guitar it's typically used in low setting.
I replaced the 50k input gain pot with a 500k pot to increase the input impedance.

Hard to tell if the broken sensitivity switch is part of the problem.
Measure it to find out what's wrong.

Thanks. I took a closer look at the schematic after I posted and saw that the high setting runs through a small transistor gain stage... so I will investigate the switch possibly being broken... but not my 1st concern obviously.

With testing, I fully reassemble, but it was sitting on top of the amp. And I did mean "acoustic feedback" - sorry, a lack of proper terminology there. And I can get crazy sounds standing close to the amp with the guitar and always stand back to eliminate that part.

I can scope it... question about that (scoping is still a bit new to me) - looking at the schematic attached below... at the bottom of the schematic are over a dozen small drawings denoted by letters that correspond to several points throughout the schematic. Am I correct to assume that I should get similar waveforms scoping these points? If that is so, point A (the input) is shown as 8mVpp - am I also then correct to assume I should feed an 8mV 440hZ signal here? 8mV seems awfully small.

Previously I did also have a 500k post installed in this... I wanted to set it back to original because I saw somebody use theirs with a buffer in front and it sounded great... plus I wanted to keep some of it's tonality... so I am still on the fence with that - I will determine what do to about that part next.

Thank you, as always!

[Helmholtz]

... at the bottom of the schematic are over a dozen small drawings denoted by letters that correspond to several points throughout the schematic. Am I correct to assume that I should get similar waveforms scoping these points? If that is so, point A (the input) is shown as 8mVpp - am I also then correct to assume I should feed an 8mV 440hZ signal here? 8mV seems awfully small.

It seems the signal levels shown are for the high sensitivity/mic setting.
In low sensitivity, input signal should be around 90mV. Adjust input signal and level control to get 4Vp-p at point C.
Signal needs to be square-wave.

But I would start looking for signs of oscillation at the output first.

Previously I did also have a 500k post installed in this... I wanted to set it back to original because I saw somebody use theirs with a buffer in front and it sounded great... plus I wanted to keep some of it's tonality... so I am still on the fence with that - I will determine what do to about that part next.

With a buffer in front the pot value is non-critical.
Without it the 50k pot will load down the PUs, reducing treble and level.

In other words, with a 500k pot the sound should be the same as with a buffer.

[Gtr0]

It seems the signal levels shown are for the high sensitivity/mic setting.
In low sensitivity, input signal should be around 90mV. Adjust input signal and level control to get 4Vp-p at point C.
Signal needs to be square-wave.

But I would start looking for signs of oscillation at the output first.




With a buffer in front the pot value is non-critical.
Without it the 50k pot will load down the PUs, reducing treble and level.

In other words, with a 500k pot the sound should be the same as with a buffer.

Thank you, Helmholtz - I don't know what I'd do without your help... on many topics! I will look at my signal generator - though I don't recall being able to set up for a square (which I see in the 1st image is a square wave), but it does have a ton of options.

Thanks!

[Gtr0]

Well, I have done some scoping. And this is where my lack of experience hurts as I am not sure of how to acquire anything other than the most basic waveforms. I adjusted C to be exactly 4Vpp which was a result of sending a 90mV square from my signal gen. and turning the input trim on the CE-1 to about 3 o'clock. C and D were exact. As for after that? It seems some are correct and others are not... but I have a feeling it's more that I do not how to properly use my scope. I have since busted out the manual and will look at some refresher videos.... part of the problem is that I need my scope like once per year, so I have never quite dug too deep into how to use it... sadly.

Signal generator was set to 90mV, square wav at 440Hz

I have attached some images.
At screen "E" I assume I need to somehow capture 3ms. I am reading the manual for that.
Screen G looks right, though the arrows indicate that the wave was changing width depending on the chorus setting (which was at min for all shots, vibrato at max)
Then at screen I, it heavily deviates from the schematic images. Same with images K and L.

I will read more on properly using my scope... I am using a 10x probe.

Thanks!