About to build a Tonebender MKI. Any advice?

[LucifersTrip]

when you get owned, start insulting...very mature.

[Arcane Analog]

when you get owned, start insulting...very mature.

Please. Owned? All you do is quote other peoples work from various forums and repost voltages and purport yourself to be an expert. 

Who starts out giving advice by saying things like "after building many great sounding pedals I have this to say."

Though, after building a bunch that sounded excellent, I was going to post a "guide" with some explanations...

[Pyr0]

Will you pair just give it up 

[Arcane Analog]

Will you pair just give it up 

Agreed. You are right Pyro. Just a stupid thread now.

[Pyr0]

It would be a great thread if all you seasoned MKI builders would just tell us some ideal transistor Hfe and leakages values to use. 

[mac]

It would be a great thread if all you seasoned MKI builders would just tell us some ideal transistor Hfe and leakages values to use. 

There are some guidelines for Q2 and Q3, but not a precise answer for your question. This is because of the very nature of germs.
Tiny changes of hfe, leakage or even Vbe from device to device, and temperature of course, have big impact on the collector voltages. What you heard yesterday while breadboarding can, and will, sound differently next day, if you live like me by the sea, or in desert areas
Some weeks ago, I put a set of leaky 2sd352 matsushita germs, +600ua, hfe:65, and it sounded great with stock values. But it was hot the first day. Next days temperature went down by 15C and both Q2 and Q3 went up and sounded more bizarre, not bad but different.

I said nothing new, I know, but I hope this example can clarify things a bit for the one time builder.

The stock circuit calls for leaky germs, 500ua<leakage<1000ua, 50<hfe<150.
Those eBay ACXXX many members complain about can be OK for this.

So if you plan to build this for your own use, and you have just a few germs that are non-leaky, or for any reason do not bias ok, you'll have to tweak/add some resistors.
Which ones?

Q1
If emiter voltage is below a volt because your germ is low leakage, some mega ohms resistance from Vcc to B will increase emiter voltage.
My advice is to set Q2 emiter above 2v.

Q2
The problem here is to make the fuzz pot work all the way up and down. If leakage is low, the fuzz pot will set collector voltage near 9v until it reaches say 10k or more. If this is the case, increasing the 1k8 resistor to 4k7 or more can help.
Q2 collector should vary from Vcc to 5v - 6v or so.

Q3
The 8k2 sucks a lot of leakage from the base, that's why you need a leaky germ here.
Non-leaky germs will hardly bias here. In this case you can send collector voltage down by adding a resistor +470k or so, from Vcc to B, to feed the base.
Some like Q3 collector ringing near Vcc, some like it at 7v, others at 4.5v. Salt and pepper to taste.

Would this changes have a great impact on the sound by changing impedances?
Q2 and Q3 input impedances are very low for this changes to have great effect.

For this reasons I designed a Silicon TB MKI 

I think it would be a nice idea if we all work on better biased GE TB MKI, designed for a specific range of germs to get consistent results.

Now, how companies build this pedal without using trimmers that are a waste of time and money?
By wasting time and money making deals with suppliers to get devices on a specific range.
I'd really like to compare many pedals from the same company and check if they all sound the same.

mac

[superferrite]

Good God.

It's no wonder that I haven't tried the Tonebender Conundrum yet, after building for over 5 years.  And I LOVE the sound of this circuit.  Or at least some of the sounds, it seems.

Thanks for the good advice nonetheless.  Mac, I for one hope you can get that Silicon Mk I to sizzle just right.  The desert is not kind to Ge pedals.

[Pyr0]

Thanks mac, some good info there. I'm also following progress in your silicon MKI thread. It's starting to sound good.

[jrod]

Hey guys!

Check out this video and listen to the noise it makes at around from about 3:10 to 3:23 (it's especially bad at about 3:20).

This is exactly the noise I get with the circuit on the breadboard as the notes decay. Is this normal for this circuit and is there any way of getting rid of it?

I am assuming that biasing the transistors to gate will help. Also, I can get rid of the noise by using a low gain transistor for Q1 (about 30-40 hfe), but then the fuzz is too light.

I have had a pretty easy time getting good voltages with the stock circuit, but that noise is awful!

Otherwise, the circuit sounds great!

[mac]

I never had that noise, even when transistors, Si or Ge, were very close to Vcc.

mac

[jrod]

Thanks, Mac! Any idea on what could be causing it? I thought it was just me until I found the example I posted.

[Electric Warrior]

I had that kind of noise. Swapped the 180k between Q2's base and -9V for a 470k and all was good. I used OC75s btw..

[jrod]

Thanks, EW! I'll give that a try! I am also using OC75's, which so far are my favorite in this circuit and the MKII.

[Arcane Analog]

This is exactly the noise I get with the circuit on the breadboard as the notes decay. Is this normal for this circuit and is there any way of getting rid of it?

I am assuming that biasing the transistors to gate will help. Also, I can get rid of the noise by using a low gain transistor for Q1 (about 30-40 hfe), but then the fuzz is too light.

I have had a pretty easy time getting good voltages with the stock circuit, but that noise is awful!

Otherwise, the circuit sounds great!

This is pretty much case in point as to what I have been saying all along in this thread.

The crackling fizzles on the deacy are very common - if not the norm. The circuit is either misbiased or, if you are seeing close to the voltages you should, you need a new transistor(s). I have said it a bunch in this thread - even with proper voltages some transistors just sound like that. No amount of trimmers/voltage corrections will get rid of it. Anyone who has built several of these should know this to be a fact and a hard reality. You can minimize the number of transistors that will not inherently work by using OC75s and AC125s - they seem to work well in this circuit. Other folks like AC128s but I have never tried any as the OC75/AC125 combo works like a charm for me.

In my experience, a minimum HFE ~70 is required for Q1. Anything less and you will probably not have enough fuzz.

Again, I have said it many times in this thread, getting the voltages where you want them is pretty damn easy. Trimmers can certainly get you there. What the argument between myself and Lucy has been about is that I contend that trimmers cannot always make the transistor sound good or remove the crappy artifacts. You can play with trimmers until you are blue in the face and utterly insane and never get a good sounding pedal. Or, once you have determined that your tranistor is biased properly but still sounds like crap only swapping different transistors can help you out. Simply starting with a pile of transistors is always going to be your best bet. Some OC75/AC125 combos are plug and play with stock values. Indeed, I have never built this circuit with anything but stock values (except the 180/470K swaps or 'Hurst' values) and there are a great number of others that have had success doing the same.

[jrod]

Thanks, AA! I appreciate what you are saying. Unfortunately, I don't have a large stock of OC75's to work with (I have 9 or 10) and the 6 AC125's are all a bit high beta for this circuit (low 200's). However, I do have a lot 2N13XX and have swapped transistors for hours and always had the noise.

Like I say, the only thing that got rid of the noise was low gains for Q1 and you are right, the fuzz is lacking.

I did just listen to the blondegraemey video that LT posted and his use of the trim pot seems to kill the noise, which leads to me to believe that the noise issue stems from Q3 bias. Maybe Q3 is the one to focus on swapping out?

I'm getting married this weekend, so after the dust settles, I am going to get back to this circuit and try these suggestions.

[Arcane Analog]

Sorry - I thought you had stated you had good voltages on the circuit including Q3. If the voltages are off it will probably sound off.

Swapping out Q3 for me is a preference for highend content. Higher HFE = more treble. Lower leakage makes for lower noise.

This circuit is all over the place though. I have had combos work that shouldn't.

[jrod]

Sorry - I thought you had stated you had good voltages on the circuit including Q3. If the voltages are off it will probably sound off.

No, you are correct. I don't seem to have a problem getting the "right" voltages. It's just that that dreadful noise I get on the note decays. The circuit sounds great until the notes start to trail off. My experience so far has been, no matter what combo I have tried, the noise is always there. I was just wondering if adjusting the bias on Q3 would help get rid of the noise. I swapped out a bunch of transistor combos, but adjusting resistor values is something I haven't done because I want to use the stock circuit.


Thanks for your input, AA!

[Arcane Analog]

You can check videos of some 'commercial' builders who send them out with the rough decay.

The thing with this circuit is that no one builds a great number of these for good reason. They arebpretty much only a commissioned pedal.

[jrod]

Yeah, I've wondered if the key to using this circuit to just not let the notes ring out too long! 

[mac]

Thanks, Mac! Any idea on what could be causing it? I thought it was just me until I found the example I posted.

Mhhh, hard to say. When do you get that noise, near Vcc? Maybe one transistor is too off. In that case try to turn it on a little more.

I had that kind of noise. Swapped the 180k between Q2's base and -9V for a 470k and all was good. I used OC75s btw..

Or the 470K for a 180k?

mac