Leyland Pedals Ultra Simple Tremolo layout and some questions...

[moid]

Hello everyone

Yes I know I should be fixing that amplifier first, (I'm working on it) but at the moment I've managed to damage one of my knees* and I'm on pain killers and I don't trust my somewhat medicated brain around a 240V circuit that I already shocked myself on once while sober, so I thought, aha, let's work on something easy and 9V instead! That way I can't hurt myself when the circuit invariably goes wrong...

I found a layout a few months ago posted by the chap who runs Leyland Pedals and he called it the Ultra Simple Tremolo, and it has very few parts and I thought I'd build that (I'd also like to rebuild some confidence; I have only built three pedals in the last 12 months and none have worked). His layout is below:



So I've laid this out in DIYLC on vero and I wanted to check if it is right before I build it - I'm not sure if I've drawn the two 10uF capacitors the right way round. I think the rest of the circuit is correct....




Now if the above works, then I was thinking... this circuit uses a dual op amp, one amp of which is doing nothing... and tremolos often appear to suffer from a drop in (perceived) volume, so could I use the other remaining op amp as a boost? So the below schematic is my guess at how to do this (using U1A of the TL072 as the boost). I'm not sure what value capacitor to put on it, or what value the pot should be, or even if this would work, but it would be lovely to hear anyone's suggestions.



Hope your Summer is going well!


*just because you can repair some damage to your roof for free, doesn't mean it's always a good idea to do it yourself

[Rob Strand]

A couple of things  it's a good idea to at least add a series resistor like the Dallas Arbiter Tremolo,

https://viva-analog.com/dallas-arbiter-trem-face-rare-re-build/

The transistor needs to work as part of a voltage divider.  In the Leyand it's relying on the pickup providing the impedance but it's not going to work well if you place buffered output pedals before it.

As far as your gain restoring circuit goes it needs work.   In short just take any non-inverting opamp circuit and put it after the tremollo,

https://www.electrosmash.com/mxr-microamp

No doubt you can strip it back a bit.

The main issues with your schematic are:
- The opamp connection has a lot of issues since you are DC coupling the input.
- The inverting scheme doesn't lend itself to the best solution.   (It is possible to do a inverting version
   but it will be less optimal than the non-inverting.)
- You are taking the 4.5V/Vref line from the oscillator circuit.
   It looks like a 4.5V/Vref, and it kind of is, but it's not a Vref suitable to connect to audio.
   That 4.5V/Vref is part of the oscillator circuit and works in conjunction with the 220k
   resistor feeding back to the opamp + input.   The voltage at that point is varying
   if you use it for audio it will tick.   You can't even add a cap across that Vref to fix it
   since it will stop the oscillator.

If you look at the Schaller Tremolo you can see a similar kind of idea which uses a BJT
amplifier to restore the level.   It's both similar and different.  The colorsound is another one.
IIRC the DIAZ Tremodillo was a another  transistor based tremolo.

You can stick to your plan using an opamp.

[moid]

A couple of things  it's a good idea to at least add a series resistor like the Dallas Arbiter Tremolo,

https://viva-analog.com/dallas-arbiter-trem-face-rare-re-build/

The transistor needs to work as part of a voltage divider.  In the Leyand it's relying on the pickup providing the impedance but it's not going to work well if you place buffered output pedals before it.

Thanks Rob! I had a lok at that schematic and I was not sure what the series resistor was - from comparing the Leyland with the Dallas, my guess is that it is the 10K on the Dallas that is after the audio in - I've added it below and labelled it 10K series... but then I saw you say it needs to be a voltage divider, so would that mean the resistor needs to move to the right (past the junction with the 2n5088?) and also should it be the same value as the 1M resistor that goes to ground after the audio input ( I am assuming that would be the second resistor in the voltage divider? I think a voltage divider need two equal value resistors?). In short, I'm not sure (sorry!)

As far as your gain restoring circuit goes it needs work.   In short just take any non-inverting opamp circuit and put it after the tremollo,

https://www.electrosmash.com/mxr-microamp

No doubt you can strip it back a bit.

The main issues with your schematic are:
- The opamp connection has a lot of issues since you are DC coupling the input.
- The inverting scheme doesn't lend itself to the best solution.   (It is possible to do a inverting version
   but it will be less optimal than the non-inverting.)
- You are taking the 4.5V/Vref line from the oscillator circuit.
   It looks like a 4.5V/Vref, and it kind of is, but it's not a Vref suitable to connect to audio.
   That 4.5V/Vref is part of the oscillator circuit and works in conjunction with the 220k
   resistor feeding back to the opamp + input.   The voltage at that point is varying
   if you use it for audio it will tick.   You can't even add a cap across that Vref to fix it
   since it will stop the oscillator.

If you look at the Schaller Tremolo you can see a similar kind of idea which uses a BJT
amplifier to restore the level.   It's both similar and different.  The colorsound is another one.
IIRC the DIAZ Tremodillo was a another  transistor based tremolo.

You can stick to your plan using an opamp.

thanks again - that looks a lot more complicated than I envisaged... I presume this is due to the IC involved? Or maybe I can remove a lot of the components? I built a very loud clean booster using an LM386 chip last year that was very simple and had only a fraction of the amount of components that the MXR microamp has...

[duck_arse]

moidy old bean - you very need to add power supply connections to the IC - ground to pin 4 and supply to pin 8. and we NEVER leave an unattended opamp like that, even if we aren't using it, we properly terminate it.

see if you can find a copy of the vicovibe circuit diagram - it's actually a trem.

summer? when?

[ElectricDruid]

That's looking a bit better. The 10K works with the transistor acting as the lower resistor to make a voltage divider. When the transistor is off, the lower part of the divider is a very high resistance and virtually all of the signal goes through. When the transistor is on, the lower part of the divider is a very low resistance, and only a fraction of the signal goes through. Hence, the magic of volume control and tremolo!!

I did a similar thing recently with a vactrol. In fact, there's quite a lot of this that is similar, so you should take a look:

https://electricdruid.net/two-stomplfo-projects-mini-tremolo/

You could directly lift the buffer-with-gain I used at the input to fix your volume issue:



I think you'd been thinking of putting the boost after, but it could go before instead. Or drop R2/R3/R4/C5 (the usual input components) and stick it after.
Duck's right though, you need to do *something* with that other op-amp. You can't leave it floating about like that. Not good practice.

HTH,
Tom

[moid]

moidy old bean - you very need to add power supply connections to the IC - ground to pin 4 and supply to pin 8. and we NEVER leave an unattended opamp like that, even if we aren't using it, we properly terminate it.

see if you can find a copy of the vicovibe circuit diagram - it's actually a trem.

summer? when?

Thanks Duck, long time no quack! I did wonder about that - I was thinking how does this opamp work without power, and then I thought, well the power is coming in through pin 5 and the op amp isn't being used to amplify audio so maybe it doesn't matter... but it obviously does and the original schematic doesn't tell me this (presumably because it's so basic!). I will re draw the schematic and layout as soon as I am more lucid.

I will search for the vicovibe - I found a thread about it on TheOtherPlace... but there are multiple versions of the schematic there. I think this one is the circuit without any mods?


From the demos I've heard it sounds nice, but the people on the other forum reported a lot of issues with getting it to work.... was there something specific to look at on it? I did find one suggestion for adding a rate LED that I might try to see if it works with this circuit

[moid]

That's looking a bit better. The 10K works with the transistor acting as the lower resistor to make a voltage divider. When the transistor is off, the lower part of the divider is a very high resistance and virtually all of the signal goes through. When the transistor is on, the lower part of the divider is a very low resistance, and only a fraction of the signal goes through. Hence, the magic of volume control and tremolo!!

Thanks Tom! So I put it in the right place, but for the wrong reason? Sounds like the story of my life! I was thinking that the 1M and the 10K were the voltage divider, never thought to consider that the 2n5088 connects to ground as well.

I did a similar thing recently with a vactrol. In fact, there's quite a lot of this that is similar, so you should take a look:

https://electricdruid.net/two-stomplfo-projects-mini-tremolo/

You could directly lift the buffer-with-gain I used at the input to fix your volume issue:



I think you'd been thinking of putting the boost after, but it could go before instead. Or drop R2/R3/R4/C5 (the usual input components) and stick it after.
Duck's right though, you need to do *something* with that other op-amp. You can't leave it floating about like that. Not good practice.

HTH,
Tom

Brilliant thanks very much, once I can think straight I will re draw and post something using your ideas. At the moment things are a bit 'end of 2001', Dave.

[Rob Strand]

I've added it below and labelled it 10K series... but then I saw you say it needs to be a voltage divider, so would that mean the resistor needs to move to the right (past the junction with the 2n5088?) and also should it be the same value as the 1M resistor that goes to ground after the audio input ( I am assuming that would be the second resistor in the voltage divider? I think a voltage divider need two equal value resistors?). In short, I'm not sure (sorry!)

The 10k *forms* a divider with the transistor.   No need to make a divider by adding a resistor to ground, the transistor is doing that already.  Without the 10k you don't have a good divider so the tremolo action won't be reliable.  (and yes 27k is fine.)

[duck_arse]

opamp pin 5 is being biased, for signal swing. biased is not powered, powered is for supplying current, biased is for shifting signal centre. always, an opamp needs supply pins connected to a supply, it should be explicit on any and every circuit diagram, but is sometimes shorthanded as being understood. which, on the circuit diagram of a super-simple circuit, is an error. because newbs.

the vico-vibe - pffffttt! TheOtherPlace? what do they know? breadboard it, see if it works. if it doesn't oscillate, try with a phase shift osc instead. or that opamp oscillator you've got in the leyland. british leyland? it'd go on strike, then rust.

[moid]

Hi chaps

Thanks for the comments. I will try to reply tomorrow, my brain is very fogged at the moment, my knee looks like the football in Duck's avatar. I've redrawn the schematic with ElectricDruid's boost in it, I think it's correct... hope to be more compos mentis or generally amusing tomorow

[ElectricDruid]

Not bad, but you've gone and used that LFO divider as Vref again. Rob warned you against that earlier on. You'll have to use a couple more resistors and an electro cap to make a proper Vref just for the booster amp. Say 10Ks and 10uF.

Otherwise it looks good now.

[Rob Strand]

Not bad, but you've gone and used that LFO divider as Vref again. Rob warned you against that earlier on. You'll have to use a couple more resistors and an electro cap to make a proper Vref just for the booster amp. Say 10Ks and 10uF.

Yes, using the oscillator "Vref" is a problem - in my eye's it's not even a Vref as the voltage there intentionally changes.   You need a separate Vref for the audio.

[PRR]

> make a proper Vref just for the booster amp.

And apply it to a cap-coupled "in" pin. (The "-in" pin can go to ground through a cap.)

[moid]

The 10k *forms* a divider with the transistor.   No need to make a divider by adding a resistor to ground, the transistor is doing that already.  Without the 10k you don't have a good divider so the tremolo action won't be reliable.  (and yes 27k is fine.)

Thanks Rob... I'm not sure what the 27K is though, I  don't think I put one on the schematic?

opamp pin 5 is being biased, for signal swing. biased is not powered, powered is for supplying current, biased is for shifting signal centre. always, an opamp needs supply pins connected to a supply, it should be explicit on any and every circuit diagram, but is sometimes shorthanded as being understood. which, on the circuit diagram of a super-simple circuit, is an error. because newbs.

the vico-vibe - pffffttt! TheOtherPlace? what do they know? breadboard it, see if it works. if it doesn't oscillate, try with a phase shift osc instead. or that opamp oscillator you've got in the leyland. british leyland? it'd go on strike, then rust.

Thanks Duck - I just found the original post on reddit where the circuit creator was helping people make this tremolo and a few of them didn't realise that the opamp needed power - it's here in case any wants to read it, and there's a video in it showing the tremolo in action as well

https://www.reddit.com/r/diypedals/comments/kgs9ki/ultra_simple_tremolo_schematic_and_video/

What's a phase shift oscillator? Sounds interesting but I think I'd better get one pedal working before starting another (glances at plastic box full of unfinished vero constructions). Did British Leyland have a bad rep in Australia as well? I would've thought with the drier climate there that they would've been better off? They certainly rusted in the UK, or at least the vehicles they made in the 70s and 80s did; my dad used to drive one for one company he worked at and was never complimentary about them.

Not bad, but you've gone and used that LFO divider as Vref again. Rob warned you against that earlier on. You'll have to use a couple more resistors and an electro cap to make a proper Vref just for the booster amp. Say 10Ks and 10uF.

Otherwise it looks good now.

Thanks Tom - it looks like I didn't understand that then! I thought (probably wrong!) that VREF was half the normal power going to a circuit, which is what I connected to... so I guess that's wrong... or maybe it's right, but I think Rob and PRR are explaining that I should have two voltage dividers because using the first one twice might cause issues?

Yes, using the oscillator "Vref" is a problem - in my eye's it's not even a Vref as the voltage there intentionally changes.   You need a separate Vref for the audio.

Thanks Rob, presumbly because a VREF needs to be a stable contstant value, and the one used here can fluctuate?

> make a proper Vref just for the booster amp.

And apply it to a cap-coupled "in" pin. (The "-in" pin can go to ground through a cap.)

Wow thanks very much PRR! Can I draw a vero layout from this schematic now? I asume the exclamation marks are to hightlight where I've gone wrong, or where I need to make changes? I have another question (from sheer ignorance) I put the audio into the negative input of the second opamp and you've changed that to the positive input - I thought (from looking at other circuits) that the audio went into the negative and the pot and other components went to the positive input - so presumably they can be swapped and it doesn't matter, or they need to be swapped in this instance?

[moid]

Another question (sorry!) The original creator (in his reddit thread) mentioned that that circuit could be modified with a capacitor inbetween the 2N5088 emitter and ground to form a low pass filter effect which sounds rather fun and presumably I could add that on a SPDT switch with the output of the emitter going to the central pin of the switch and then one of the other pins going to ground and the other also going to ground by via a small capacitor like say a 80 - 100nF? Like the below - or would this cause something else in the circuit to go wrong?



He also mentioned that a high pass circuit could be made, although I doubt that would be possible without major changes to the first part of the circuit. I'm now wondering whether I could put a pot in place of the capacitor and have the emitter go to the wiper of the pot and then have two different cap values on lugs 1 and 3 of the pot, both going to ground to allow the low pass filter to have its cut off point adjusted? Sorry if I'm rambling, I am not 100% with the planet at the moment due to painkillers (don't do drugs and electronics kids!). My knee looks less inflated this morning so that's good!

[duck_arse]

one input will invert the signal out (- in) and the other will not invert the signal outputted (+ in). how each input is used and connected for gain/s is a little different. a topic for research.

aboot yer cap to emitter. yes, like that, most probably, I haven't tried it, but you will. and, for your differents cut-offs filter attempt - get a single pole double throw centre-off switch. [or double pole if that's all you have.] wire your smaller value cap from emitter to ground. connect switch common to emitter//small cap. centre position therefore equals small cap. connect one end/throw of switch to ground. connect other end/throw to a larger value cap, and the other end of that cap to ground. now when you switch one way, it shorts the cap and works stock standard. when you throw other way, larger cap is in parallel, and your filter shifts downwards. see?

[PRR]

> VREF needs to be a stable contstant value, and the one used here can fluctuate?

It DOES fluctuate! A lot. About 1/3rd of the supply voltage. That's how the oscillator oscillates.



> I put the audio into the negative input

That's usually less-good. Here it leads to low input impedance and excess hiss. Non-inverting is really very common.

[Rob Strand]

Thanks Rob... I'm not sure what the 27K is though, I  don't think I put one on the schematic?

Early on in reply #5 you posted a circuit with 27k's.

Anything from 10k to 100k will work and trades off pickup loading vs noise.
Something like 22k might work a little more consistently than 10k with different pickups.

If you raise the series resistor you technically should increase the base resistor
on the transistor.   The way your circuit works has quite low base resistors and relies
on backing off the voltage.   Large resistors may sound a little different, perhaps
smoother.

[moid]

aboot yer cap to emitter. yes, like that, most probably, I haven't tried it, but you will. and, for your differents cut-offs filter attempt - get a single pole double throw centre-off switch. [or double pole if that's all you have.] wire your smaller value cap from emitter to ground. connect switch common to emitter//small cap. centre position therefore equals small cap. connect one end/throw of switch to ground. connect other end/throw to a larger value cap, and the other end of that cap to ground. now when you switch one way, it shorts the cap and works stock standard. when you throw other way, larger cap is in parallel, and your filter shifts downwards. see?

I thinks so... maybe? Let's see if I can translate your words into pictures! Using only the power of my mind! (whirring, ticking, mechanical noises, some puffs of steam and the hamsters start running in their wheel)



(my best second rate end of the pier magician act pose) "Is this the card you pulled from the deck?"

If I've drawn that correctly then that seems like a devilishly clever use of a switch! I've only got a DPDT On off On switch, so  I assume I just use only the lugs on one side of it and that will be fine? If it only works with a SDPT version then I will order one.

Hmmm that gives me ideas (probably not a good thing) about a high pass version... edit: nope, doesn't look like that is possible... oh well.

The reason for using a switch instead of a pot to control going between two capacitor values is because I would presumably also need a switch to activate / deactivate the pot, and also the resistance value of the pot might upset the voltage divider made by the 10K series resistor and the transistor?

[moid]

> VREF needs to be a stable contstant value, and the one used here can fluctuate?

It DOES fluctuate! A lot. About 1/3rd of the supply voltage. That's how the oscillator oscillates.



> I put the audio into the negative input

That's usually less-good. Here it leads to low input impedance and excess hiss. Non-inverting is really very common.

Brilliant, thanks PRR, I will do as shown on your schematic. Your big wobble, small wobble notes made my brain revert to the eighties - Big Bubbles, No Troubles

Thanks Rob... I'm not sure what the 27K is though, I  don't think I put one on the schematic?

Early on in reply #5 you posted a circuit with 27k's.

Anything from 10k to 100k will work and trades off pickup loading vs noise.
Something like 22k might work a little more consistently than 10k with different pickups.

If you raise the series resistor you technically should increase the base resistor
on the transistor.   The way your circuit works has quite low base resistors and relies
on backing off the voltage.   Large resistors may sound a little different, perhaps
smoother.

Found it now! blame that on painkillers (or stupidity, whichever seems more relevant!) Thanks for explaining.