Why does a Tone Bender MkII sounds good? half a cent analysis...

[Steben]

Hi guys... part 2... for part 1 click here: https://www.diystompboxes.com/smfforum/index.php?topic=125395.0
The Tone Bender Mk2.
Usually described as "more gain yet darker than a Mk1"

Base circuit:


Very important here is the use of the first stage as a common emitter stage, in other words as a gain stage instead of a buffer as in the Mk1.
Yes, more gain, but because of the topology and the absence of an RE, the input impedance implodes. With silicon standards, we get at 1k8, yet internal resistance of Germaniums are less so we get easily down to 1k2. This is heard in the treble roll off typically found in Fuzz Faces as well but even stronger present.

The current in Q2 is far less, which makes for a far higher internal resistance. With silicon numbers we would have up to 2600 ohms, but this is germanium.
Ic is somewhat around 0.1mA (100µA), yet the leakage current needs to be added. some Q's actually achieve 200µA! This means the Ie current can be just a bit more to double or even more!
This means with lower Germanium numbers, hFe range and leakage internal resistance can be ranging from 1k to 175 with ease.
This results in Zb ranging from 80k to 10k. This is a huge gap! In parallel with the feedback loop which compensates the gap we see a total input impedance of the second stage from 5k to 10k.
And this in combination with the coupling cap results in a -3dB frequency ranging from 159Hz to 318Hz

As an average at 7k for Q2 stage we see the following -3dB frequencies:


Going back to the input stage we have guitar loading.
Inductance for single coils is a about 2H average.
Input response with guitar load:


This means a -15dB level at most to already -25dB at 2.5kHz.

We can conclude the first stage compensates the loading yet with treble loss.
Combined with the low roll off at Q2, this pedal has a focus on midlow frequencies around 300Hz.
Goes without saying that the gain, bias and filtering is very device dependent. Easy to see how many pedals sounded different.

[Chugs]

Interesting.

What happens to the HPF calculations if you change the inout transistor base resistor to 10K and the second transistor resistor to 47K? (These values are more commonly seen on Tonebender MKII's.)

[Steben]

Interesting.

What happens to the HPF calculations if you change the inout transistor base resistor to 10K and the second transistor resistor to 47K? (These values are more commonly seen on Tonebender MKII's.)

Q1
Zb is about 1k2. This means with 100k that Zin remains about the same.
With 10k, we have Zin = 1/(1/10k + 1/1k2) = +/- 1k1 which is almost the same. So yes, 10k does not make a huge difference because Zb is already so low.

Q2
with 47k, IE will be higher, resulting less internal resistance and therefor higher corner frequency up to 500Hz with high leakage. Such MkII's should sound a bit less dark compared to the 100k ones. But none can sound bright mid grainy as a MkI.

MkII is designed in such a way it eventually sounds as between a Mk1 and a Mk1.5 (which is a proto fuzz face)

[Chugs]

I like this kind of circuit analysis.

What is the output impedance of the first stage?

[Steben]

I like this kind of circuit analysis.

What is the output impedance of the first stage?

+/- Rc // RL(oad = input impedance second stage)
+/- 10k // 7k = 1/(1/10 + 1/7) = 4,12 k

[hans h]

Hi Steben,

Nice analysis! Do you also have a graph for the resulting frequency response hitting the drive stage (centered at 300 to 500 Hz) ? That would be nice stuff for people trying to emulate sounds with a wholly different schematic. For example:

I recently made an opamp schem with three inductors in the feedback loop of the first stage that allows me to adjust what frequencies hit the drive stage. Would be interested to try a tonebenderish frequency response as well. At the moment the inductors are centered at 300 hz, 1khz and 3khz. I can see what is happening in a ltspice Sim of my schematic.

Thanks in advance, Hans

[hans h]

BTW I could do it myself in ltspice. Regarding freq resp, is it sufficient to use the stated zins in combination with cap values, but use opamps instead of bjt's (with same gain per stage)? I ask this because I do not have any experience with bjt's in ltspice.

[antonis]

I ask this because I do not have any experience with bjt's in ltspice.

Do not feel  embarrassed 'cause LTspice hasn't any exprerience with BJTs, also..

[amptramp]

Part of the reason this sounds good is that the harmonics of the input signal are not exact integer multiples of the fundamental due to variations in damping of the strings with frequency and fret position.  The removal of input harmonics and their replacement with exact harmonics sounds better because there are no intermodulation beat notes between the incoming harmonics and the exact harmonics the circuit generates.

The performance of this circuit changes with hfe and therefore with temperature since that affects hfe.  Since it sounds good with lower hfe values than normal, this would work well in a temperature-controlled chamber using a Peltier cooler that would stabilize the characteristics over temperature and also reduce the leakage with germanium transistors.

[antonis]

this would work well in a temperature-controlled chamber using a Peltier cooler that would stabilize the characteristics over temperature and also reduce the leakage with germanium transistors.

After constant voltage and constant current configuratios we pass to constant temperarute ones..

[Electric Warrior]

The whole point of this circuit probably was to fix the temperature issues the MK1.5 suffered from. Most were built early in 1966 in winter or spring. They were biased rather hot and will sound farty very quickly. This sure would have been a problem in summer. It's probably the reason they upgraded MK1.5s to MKII specs by adding some parts (hence the "small board" variant).

The MKII on the other hand can sound great in warm weather. They do gate more as the temperature increases, but the gating is awesome. Some work well even beyond 30°C. They do eventually crap out, but they're well behaved in a rather wide range of temperatures.

[amptramp]

You may have a gig in Nome, Alaska followed by an outdoor gig in Reno, Nevada.  You would have to have separate units or a unit where every resistor was a trimpot to make this work.  Say what you want about op amps and feedback circuitry, they were designed just to prevent this sort of thing.  People have internal temperature control, that's why we have drummers who can maintain a beat at all temperatures.  Your effects units should have the same.

[cspar]

After constant voltage and constant current configuratios we pass to constant temperarute ones..

I have an old Panasonic radio that I am pretty sure uses a thermistor for temperature compensation. It's my understanding that it's not uncommon in later germanium radios but haven't been able to find much on the technique.

I've already recapped and Deacyed the radio. It's on a really tightly packed board that's double sided and has a ton of vias. I didn't trace it, don't have the schematic and have decided against depopulating the board to trace it.

The thermistor is typically thru hole mounted and not attached to a transistor like in Jose's thermistor example in the Benson Germanium Fuzz topic.
https://www.freestompboxes.org/viewtopic.php?p=284680#p284680

Is it something along these lines what you're suggesting Amptramp? It kind of sounds to me like your suggesting making the enclosure a refrigerator/beverage cooler and not just encasing the transistor similar to Jose's example or the Benson Germanium Fuzz.

[antonis]

Is the "other forum" clearly displayed or do I need some cups of black coffee..??

[mac]

I have an old Panasonic radio that I am pretty sure uses a thermistor for temperature compensation. It's my understanding that it's not uncommon in later germanium radios but haven't been able to find much on the technique.

I've already recapped and Deacyed the radio. It's on a really tightly packed board that's double sided and has a ton of vias. I didn't trace it, don't have the schematic and have decided against depopulating the board to trace it.

My old National RQ445 germ has a kind of Fuzz Face preamp but no thermistor because Q2 emitter and collector resistors are almost the same, and so Q2C is almost fixed.
It has one in the power amp stage though, as my old Crown CRC530 silicon. Very Deacy, BTW. I use both since I was 6-7.

mac

[amptramp]

After constant voltage and constant current configuratios we pass to constant temperarute ones..

I have an old Panasonic radio that I am pretty sure uses a thermistor for temperature compensation. It's my understanding that it's not uncommon in later germanium radios but haven't been able to find much on the technique.

I've already recapped and Deacyed the radio. It's on a really tightly packed board that's double sided and has a ton of vias. I didn't trace it, don't have the schematic and have decided against depopulating the board to trace it.

The thermistor is typically thru hole mounted and not attached to a transistor like in Jose's thermistor example in the Benson Germanium Fuzz topic.
https://www.freestompboxes.org/viewtopic.php?p=284680#p284680

Is it something along these lines what you're suggesting Amptramp? It kind of sounds to me like your suggesting making the enclosure a refrigerator/beverage cooler and not just encasing the transistor similar to Jose's example or the Benson Germanium Fuzz.

Hey, if you add a beer cooler to a pedal, the guitar=playing world will beat a path to your door.

As for your other question, a number of transistor radios from the 1950's had a thermistor from base to ground in the push-pull output stages so the bias could track the variation in Vbe with temperature.  Some others used a forward-biased diode from base to emitter supply to perform the same function.

[duck_arse]

Is the "other forum" clearly displayed or do I need some cups of black coffee..??

no longer a rite of passage. brought down from on high. have as many coffeess as you see fit. freestompcoffees.com

[cspar]

Is the "other forum" clearly displayed or do I need some cups of black coffee..??

It absolutely is. Ironicly, the thread on this forum where the Benson Germanium Fuzz is independently traced and posted doesn't have Jose's neat tidbit describing a peticular thermistor technique.
https://www.diystompboxes.com/smfforum/index.php?topic=128023.msg1230725#msg1230725

After a few cups of black coffee it became apparent that I'd never searched for "thermistor" in this forum and found more information in general than over "there" including this thread with a post of yours Amptramp, which elaborates on some of your temperature control musings.
https://www.diystompboxes.com/smfforum/index.php?topic=126695.msg1221599#msg1221599

I live in Arizona. All my germanium fuzz need trim pots and/or sockets and am intrigued by temperature compensation techniques.

Hey, if you add a beer cooler to a pedal, the guitar=playing world will beat a path to your door.

You could put fuzz on effects loops in a mini fridge with a pony keg in middle of the stage... 

My Panasonic is a R-1551. It seems that the thermistor in it is for the 2SB176 PP pair but like I said it has a ridiculous amount of vias and is tight packed. All the resistors are standing and close. I tried to trace it when I recapped it but I would have had to depopulate the whole board pretty much so I didn't.

It's fun paired with a Basic Audio Scarab just using the stock speaker. Not quite a mk2 but close 

[pacealot]

I had seriously considered trying to add a Peltier beer cooler-type gadget to the pedalboard I'm glacially slowly working on, permanently installed underneath where I put my Ge fuzzes, so I could swap them out as needed but still have the advantage of keeping them cool; but I've realised that it's far beyond my limited skill set, seeing as how I still struggle with just the fuzz circuits themselves from time to time. 

But also I concur with EW on the MkII (or Supa Fuzz in my case) still sounding good even when the temperature goes up enough to throw the "normal" bias out of whack — up to a point, at least (after over an hour in the direct sun on a warm summer day, it does start to degrade in a non-musical way somewhat, but it really does sound amazing when it just starts to gate out before that point)...

[pinkjimiphoton]

when i still used ge fuzzes, i'd carry an embossing tool <heat gun> for if they got too cold... pull the input jack, and blast some hot air in there... and if they got too hot, i'd stick 'em in the club's fridge or freezer for a bit.

i don't miss them days at ALL. lol