Reducing Signal Voltage for FX Loop

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
V3a is being heavily overdriven, and is intended as a common cathode stage.
Well, that opens up another can of worms. Do you want the heavily overdriven V3a anode signal to go to your FX send? The signal at V3a cathode is probably going to be less overdriven, lacking the distortion created by V3a itself. :(

If that turns out to be the fly in the ointment, as I suspect it will, then we have to start re-thinking the problem all over again.

If you have a signal that's more than 100Vpp on the anode (i.e. more than 30 volts RMS), and you need about 100 mV RMS for the FX send, that's roughly a 300:1 attenuation.

This means you could use something like a 1 meg resistor and a 3.3k resistor to divide down the signal. 1 meg will barely load the 12AX7, and 3.3k source impedance at the FX send should be quite low enough for a valve guitar amp.

If you insist on less than 3.3k source impedance at the FX send, your best option is beginning to look like a MOSFET. Easy, cheap, no extra valves or valve sockets or heater power necessary, and you can buffer that 3.3k send impedance down to a few tens of ohms.

Personally, I wouldn't bother...IMO, 3.3k send impedance is just fine. (So is anything up to maybe 10k.)

-Gnobuddy
 
Well, that opens up another can of worms. Do you want the heavily overdriven V3a anode signal to go to your FX send? The signal at V3a cathode is probably going to be less overdriven, lacking the distortion created by V3a itself. :(

If that turns out to be the fly in the ointment, as I suspect it will, then we have to start re-thinking the problem all over again.

If you have a signal that's more than 100Vpp on the anode (i.e. more than 30 volts RMS), and you need about 100 mV RMS for the FX send, that's roughly a 300:1 attenuation.

This means you could use something like a 1 meg resistor and a 3.3k resistor to divide down the signal. 1 meg will barely load the 12AX7, and 3.3k source impedance at the FX send should be quite low enough for a valve guitar amp.

If you insist on less than 3.3k source impedance at the FX send, your best option is beginning to look like a MOSFET. Easy, cheap, no extra valves or valve sockets or heater power necessary, and you can buffer that 3.3k send impedance down to a few tens of ohms.

Personally, I wouldn't bother...IMO, 3.3k send impedance is just fine. (So is anything up to maybe 10k.)

-Gnobuddy

After some play testing, the tone is different coming off the cathode (as per the recent schematic) than off the anode. The overdrive is similar, but tonally it lacks the same punch as when the signal is taken from the plate. So yea, that idea is a no-go.

As an aside, I'm not married to the preamp circuit. I wanted to experiment with 5 cascaded stages, rather 4 (+CF) which seems to be an industry standard. Point being, nothing is set in stone and I'm willing to try just about anything. My only limitations are physical: room on the turret board, # of preamp valves, etc

The 1M/3.3k divider would work then. I didn't consider something like that because I'm still trying to fully comprehend impedance as a whole. I think I'm starting to "get it" though.
 
I wanted to experiment with 5 cascaded stages, rather 4 (+CF) which seems to be an industry standard.
I'm not much of an amp historian (I'm more interested in new ideas than old ones), but as far as I know, some early guitar amps used a single pentode gain stage, followed by the output valve.

Then the 12AX7 arrived, and Leonidas started using two gain stages, both conveniently in one cheap package.

Then we had Dumble and Ken Fischer stick in a third gain stage, and both became famous for their amps.

Somewhere along the way buzz-saw distortion arrived, and somebody (Randall Smith /Boogie Mk I?) stuck in a 4th gain stage.

I think the Peavey 5150 has five gain stages - I've seen at least one schematic showing them, I think for an Eddie Van Halen version of the 5150.

There was a monstrosity called the Bogner FISH preamp that contained eight 12AX7s, or sixteen triodes. It had multiple channels, and I don't know exactly how many gain stages there were in the nastiest of the channels. But I believe Alice In Chains was one of the groups that used the FISH.
I'm still trying to fully comprehend impedance as a whole. I think I'm starting to "get it" though.
IMO it's a pretty non-intuitive concept, particularly when it comes to things like voltage dividers. I taught myself Thevenin's and Norton's theorems when I was an eager teenager, and after that, it all made sense.

You can find those two theorems in just about any engineering textbook on electricity or electronics. There is a very short, but still useful, "Cliff Notes" version here: Thevenin's Theorem

-Gnobuddy
 
You can find those two theorems in just about any engineering textbook on electricity or electronics. There is a very short, but still useful, "Cliff Notes" version here: Thevenin's Theorem

Thanks for the link. I never had any formal training in electronics (I'm a chef by trade), and being nearly 40y/o I don't learn as quick as I once did. Not to mention real-life takes up a lot of time and energy. TBH, I wish I had such an interest when I was a kid, to pursue electronics as a career. But it's never too late to learn and I'm enjoying the process.

Back to the circuit at hand, I'm inclined to just dedicate a whole triode (output buffer, recovery stage) to the FX loop and be done with it. Design the preamp around that.
 
TBH, I wish I had such an interest when I was a kid, to pursue electronics as a career.
Be careful what you wish for, because it you may get it one day. :D

I thought I was incredibly lucky when I landed my dream job in the late 1990s, doing R&D for a successful audio electronics company. Then I found out (a) most of my co-workers were wannabe rockstars living on the hope of making it big some day, (b) most of them had been laid off many, many times from similar jobs over the years, (c) engineers have short shelf-lives: you will soon be replaced by someone younger and cheaper, and (d) consumer electronics is a luxury that nobody wants when times get tough.

Item (d) became all too clear when the dot-com collapse came along just a couple of years later. Sales plummeted, the company quickly went into the red, and then came waves of layoffs. I was one of many laid off in the second wave.

I kept in touch with some former colleagues, and so found out that there was soon a third and final wave of layoffs, followed by a former colleague's despair leading to his suicide :(, a corporate bankruptcy, and the subsequent auctioning-off of all assets and trademarks for pennies on the dollar.

After five fruitless months trying to land a similar engineering job (none of the other audio companies was doing a whole lot better in that bleak economy), I ended up taking a crap job that paid one-third of my previous income, as the alternative was to lose our house because we couldn't pay the mortgage.

Eventually I managed a successful career change, and did okay for a while. At that point my wife and I left the USA, as we could see that ugly political changes would be coming in a few years.

I'm keeping my head above water these days, but only just, and that only because I was very diligent about paying-down our mortgage as fast as I could during the years when I did have a decent income.

I hear there was a time in North American history when working as an electronics engineer was a guarantee of a good job and a life-long career. Those days are long gone.
But it's never too late to learn and I'm enjoying the process.
Agree completely. I'm still learning about electronics (particularly tube guitar amp related stuff), and still learning to be a better guitarist, and still learning to be a better musician (which IMO is quite different than being a guitarist).

In my experience, electronics is considerably more fun as a hobby than it was as a profession. What I miss is all the expensive test and measurement gear I could never afford for myself. But these days there are cheap good-enough substitutes for much of it. :)

Back to the circuit at hand, I'm inclined to just dedicate a whole triode (output buffer, recovery stage) to the FX loop and be done with it. Design the preamp around that.
That's definitely one way to skin the cat!

-Gnobuddy
 
I thought I was incredibly lucky when I landed my dream job in the late 1990s, doing R&D for a successful audio electronics company...
Man, that's quite a tale. I can relate somewhat. I've been in the restaurant business for a little over 20 years, first as a cook then eventually a kitchen manager/chef. While the work is steady (you can always find a job), it's more of a "labor of love". Meaning, the pay is crap, especially compared to the amount of work to earn a paycheck. When you're young, it's fun (fast paced environment) and exciting (plenty of drinking buddies). As you get a bit older, the 12 hour days on your feet in a hot kitchen catch up to you.

Then there's the management aspect (which is were I became burnt out). It becomes less about passion for food and more about the bottom line (it is a business after all). To the point that you have to hire mediocre staff (payroll is a huge expense) and buy lower grade products. I dreaded Monday morning conference calls with the corporate offices. It was my only day off, but I still had to go in and listen to them bitch about food cost, labor cost, and why sales were down 0.5% from same week last year.

Fast forward to early this year. I shatter my right knee cap, which required surgery and having my leg immobilized for 4 months. Obviously I had to take a leave of absence while I recovered. So I've taken that time to pick up a new hobby (tube amps). It's kept me sane and my mind occupied. I'm still not back to work - I'm able to walk, but only short distances. Being on my feet for any amount of time, or one slight "tweak" of my knee and I'm down for the count. I've had to resign from my job because of it; it's been 9 months and I'm unsure if I'll ever have the same mobility.

But hey, that's life. Just gotta roll with it.

That's definitely one way to skin the cat!

I've attached what I have in mind. I also scrapped a lot of the preamp circuit I had. I'd found myself tacking on components trying to solve one problem, but then creating another. Sometimes the simplest answer is the right one.
 

Attachments

  • 4 stage OD buffered FX loop.png
    4 stage OD buffered FX loop.png
    64.5 KB · Views: 53
But hey, that's life. Just gotta roll with it.
Indeed! And judging by my own life, and the lives of most of my friends and acquaintances, it seems the gap between what you expected from life, and what you actually got from life, starts to get wider and wider after your 40th birthday. :D

I'm very sorry to hear about your injury. :( Let's hope you eventually make a full recovery. Both physical, and financial.

Human knees are a great example of an evolutionary flaw. They are so easy to injure, and they routinely start to wear out by the time you're in your 40's. But historically you've already had kids by then, so evolution doesn't care about your disappearing cartilage and early osteo-arthritis!

I twisted my left knee badly in my twenties, clambering over slippery wet rocks near a waterfall like an idiot. There was some internal damage, and I couldn't afford proper medical care. Three decades later, that knee still hurts now and then.
I've attached what I have in mind.
Just a thought - will you be elevating the heater of V3? With the grid of V3a at +200 volts DC, the heater shouldn't be at ground (0V), otherwise there is an increased chance of damaging the heater-cathode insulation.


I believe 200V is in fact the maximum allowed Vhk, and don't forget, you will exceed that by the amount of the positive signal swing. i.e., if you have a +100V swing on the cathode, it will momentarily go to +300 volts, while the heater is still at zero volts...

Of course V3b's cathode is nearly at zero volts, so you will have to split the difference when you plan your heater elevation voltage. Maybe +100V or thereabouts.

Incidentally, signal swing at V3a cathode is the other thing I'm wondering about. In a typical common-cathode stage, the cathode is sitting at about +1.5 volts, and so you will never get more than maybe 3V peak-to-peak out of it, making it relatively safe to feed to guitar FX pedals.

But your cathode follower has the cathode at +200V DC, (assuming the 400V B+ you showed earlier is correct. That means the cathode is probably capable of swinging at least 100 volts positive, and 200 volts negative. Conceivably, V3a could put out something close to 300 volts peak-to-peak, and that would definitely fry any guitar pedal seeing that signal.

So it's very important to make sure the input signal to V3a never exceeds two or three volts peak-to-peak. V3a itself is dangerously powerful, so you have to make sure the signal is properly throttled-down before it enters V3a. I'm sure that's why you introduced R57 / R59. Good call!

One other thing: when first powered on, the grid of V3a shoots immediately to half-B+ (200V). But the heaters are cold, so there is no current through R45, and the cathode is therefore at 0 volts. Now you have 200 volts potential difference between the very closely spaced grid and cathode...and that is likely to cause an flash-over, a spark jumping from grid to cathode, which can ruin the valve.

You can prevent this with a protection device between grid and cathode. The "classic" device is a neon bulb, or you can wire a 10k resistor in series with a silicon diode (1N4007, say), and wire the pair between grid and cathode, in such a way that the diode conducts if the grid goes positive of the cathode, and the 10k resistor then limits how positive the grid can go. (See the Valve Wizard's write-up here: The Valve Wizard )


-Gnobuddy
 
Human knees are a great example of an evolutionary flaw. They are so easy to injure, and they routinely start to wear out by the time you're in your 40's. But historically you've already had kids by then, so evolution doesn't care about your disappearing cartilage and early osteo-arthritis!

I have a 13y/o son. I'm just glad I didn't have this injury when he was still a baby. That added responsibility of caring for a little one would have made recovery unbearable. He's at that age where I'm "uncool". If he only knew lol...

I twisted my left knee badly in my twenties, clambering over slippery wet rocks near a waterfall like an idiot. There was some internal damage, and I couldn't afford proper medical care. Three decades later, that knee still hurts now and then.

Ouch. No insurance is no fun when you need it. I've been fortunate to never had anything major, health wise, until my knee injury.

Just a thought - will you be elevating the heater of V3? With the grid of V3a at +200 volts DC, the heater shouldn't be at ground (0V), otherwise there is an increased chance of damaging the heater-cathode insulation.

I'm conflicted about this, honestly. I've read about it (valvewizard's site), but have yet to see it applied in practice (Meas, Soldano, Fryette, etc). Does that mean it's ok? No. But on the flip side, just because a self proclaimed wizard on the internet says it's bad juju doesn't mean you HAVE to.

Personal experience, my Triple Recto (from the 90's) uses basically the same arrangement for it's output buffer. B+ at that node is something like 395V IIRC. That's been my gigging amp for over 2 decades and never given me a problem. I'm not trying to argue with you, but until I personally experience an issue with cathode stripping I won't bother.
 
He's at that age where I'm "uncool". If he only knew lol...
I think we can put much of the blame on Disney and the power of immersive advertising, both telling every generation that they need to completely re-invent everything their parents did.

Of course, this is quite impossible. Get deep enough, and we all have mostly the same needs, the same wants, and the same fears. Even the same language and shared history. So there is actually very little of significance that each new generation can change, even if they try. The most you can do is try to change a few ugly cultural things for the better.

As a very wise woman once said, there are only three things that human beings can actually do - we can sit, stand, or lie down. Everything else is just a story.

(I would add that we can float, though most of us rarely do. :) )
...my Triple Recto...something like 395V...over 2 decades and never given me a problem.
Well, that's certainly a very solid data point!

I have never owned a guitar amp with a cathode-follower, so you have actual experience, while all I have is book and Internet information.
I'm not trying to argue with you
By all means argue with me. Nobody died and left me in charge of anything, not to mention, constructive criticism is one of the best ways for all involved to learn something new and move forward. :)

It's the destructive type of arguing - hurling contempt and insults while completely blocking out the other person's point of view - that is completely useless. It figures that certain Very Important Politicians have now adopted that useless approach pretty much all the time. :rolleyes:

Back on topic, I'm curious if your planned cathode-follower will produce yet another change in tone. Hopefully, it will be one you like.

-Gnobuddy
 
By all means argue with me. Nobody died and left me in charge of anything, not to mention, constructive criticism is one of the best ways for all involved to learn something new and move forward. :)

Yea, I'm hesitant because I'm still in the beginning stages of learning tube circuits, and electronics in general. The hardest part about being self taught is sifting through all the information out there and deciding what's the most credible. I'm in no position to debate here lol! I'm waaay out of my league. I just lurk, and post a thread here and there when I feel I need some direction.

Back on topic, I'm curious if your planned cathode-follower will produce yet another change in tone. Hopefully, it will be one you like.

I have used it before in a previous preamp build, and it seemed to be pretty neutral. From what I've read, it should remain linear unless it's overdriven. In this context, the signal will be knocked down to less than 1Vpeak before it hits the buffer, so there's no chance of that. And it should be easy to A/B the differences.
 
TBH, I wish I had such an interest when I was a kid, to pursue electronics as a career. But it's never too late to learn and I'm enjoying the process.
Well.I did :)When I was 16 I started studying Industrial Engineering and wanted a "peaceful noiseless" Hobby for fun, since I moved to Downtown Buenos Aires (think Downtown NY or London for comparison) from a rural City in the middle of the Pampas where I had enjoyed model plane building and flying, launching homemade rockets, wild stinking Chemistry experiments, etc. , much frowned upon by close neighbours, typical of living in the dense built core of a large City.

I was a big fan of the old Popular Mechanics magazine and built as many projects from it as I could, but most activities were noisy, foul smelling or needed machinery unsuitable to appartment life, so Electronics was both cool and inconspicuous.

But before the current Oriental onslaught, (we are talking 1969 on here) "whatever you made you could sell" so I started making and selling "color organs" (light show machines) to Discos, then Strobe lights (I was the first guy making them in Buenos Aires), my first Guitar product was a Treble Boost pedal (think the silicon version of Dallas Rangemaster) straight out of Popular Mechanics, then my first Bass amp: a Fet input Ampeg BT15 preamp clone driving a Motorola 100W amplifier, then a long string of Blackface Fender clones, from a Champ to Bassman 50 to Twin Reverb versions.

That started in 1969, it will be 50 years in the Profession next February, have already built and sold over 14000 amps and cabinets, (lost count around 2004) but my current customer database , started in 1999, holds some 8500 phone numbers.

Oh, and what happened with the University?
Actual manufacturing and selling took more and more of my time, plus I had to "adjust" my career to actual needs, so after 4 years of Industrial Engineering I switched career name and University and added 2 years of Electronics Engineering and then switched to 1 extra year of Business Administration.

So I don´t *really* have a Degree hanging from my wall although my formation was very solid and focused ... plus 50 years experience actually *doing* it.

In fact, former University pals, who persisted and got their degrees, generally don´t work "as Engineers", running actual production in some Industry, but do Administrative work, or are glorified salesmen ("Sales Engineers") or are registered "Expert Witnesses" at some Court or glorified Math or Physics teachers and so on.

If I had to do the same today, it would be all but impossible, who can beat *cheap* *cheap* Oriental Electronics, which to boot are rising in quality?
 
Well.I did :)When I was 16...

That's a pretty impressive resume. While it may seem impossible today, with the internet and social media you have access to a much larger market. Sure, the competition is proportionally just as big, but just look at all the "boutique" pedal companies out there now. It seems like there's hundreds of 'em making Tube Screamer clones. All it takes is a well known YouTuber to review your product and then boom - you're the next "big thing".

While it' easier said than done, so is the career path that you've managed for yourself.

While I'm in no position at the moment to do it, I've pondered the idea of starting an audio company. The pedal industry is completely saturated, but there will always be a market for tube amps. I guess it's easier entry to make pedals, which suggests why there's so many companies (likely most are just a dude and his wife) doing it. Most of the small amp companies seem to just make Fender/Marshall/whatever clones. I've seen very few that actually try to develope their own unique "sound". I guess whatever sells, right?
 
If I had to do the same today, it would be all but impossible
What you have done is pretty amazing. You are Argentina's own Leo Fender mixed in with Hartley Peavey (from the days when he was an innovative manufacturer who made really good products.)

I know those once-amazing pampas that you grew up on, have mostly disappeared now too, just like the possibility of a simple and successful life as an electronics engineer.

The electronics engineers I knew all either changed careers, became (bad) managers, or became "field application engineers" for semiconductor manufacturers.

"Field application engineer" is management-speak for "underpaid person who drives around trying to sell our parts". I still have memories of one particular "Field application engineer" who would show up in a tight white shirt and tighter blue jeans, sit on the desks of single young engineers, flipping her hair casually and flirting with them like crazy in the attempt to get them to order semiconductors from the company she worked for.

I'm not sure whom I felt more sorry for - the "engineer" flirting to try to make a living, or the lonely engineers whose vulnerabilities she was trying to take advantage of.

-Gnobuddy
 
I have been reading this thread with great interest. The way Soldano did the effects send is not really common I think.

I am interested in the output impedance of this circuit.
Is it the CF output impedance + the voltage divider output impedance or is the CF output impedance replaced by the voltage divider output impedance?

If it is the latter I wonder what would happen if the resistor to ground would be smaller then the CF output impedance ...

Hope you don't mind that I bring this up in your thread Fozzy2015
 

Attachments

  • soldano fx loop send.gif
    soldano fx loop send.gif
    4.2 KB · Views: 61
I don't mind at all! In fact, that very circuit was what originally inspired this thread. IMO, if you ignore the fact that it's DC coupled, it's not uncommon (most modern amps use AC coupled). In the SLO's instance, that particular CF is intergral to the preamp's tone, and IMO is a nifty way to drop the signal down to manageable levels. It's funny that it's often criticized - even suggesting Soldano "made a mistake". I recently saw a video of him talking about the SLO loop, and he indeed designed that way on purpose (obviously).

I don't know the exact answer to your question. I'm sure someone will chime in soon. My guess is it's 2.2k || impedance of CF.
 

PRR

Member
Joined 2003
Paid Member
The CF is "few K". This is in series with 100K, so we have 100+few K. Pessimistic, 103K.

The other path is 2.2K.

at most: 103K||2.2K = 2.154k
at least: 100K||2.2K = 2.153k

The CF impedance is nearly irrelevant.

Note that if *instead* we had come from the plate of a common cathode stage, Zout like 39K, it is 2.16k so not significantly different in impedance. (Apparently the distortion is different.)
 
The other path is 2.2K.
There is also the extremely undersized 1 uF coupling capacitor to consider. At 80 Hz, it has an impedance of about 2k ohms, which has to be added to the output impedance of the CF + resistors, essentially doubling the output impedance of the entire send circuit.

Using a 10uF cap would have virtually eliminated this problem, and 10uF caps only cost pennies more than 1uF caps, worst case.

What a pity to double the send impedance simply by using an inadequate coupling cap!

-Gnobuddy
 
So if we would make the 2.2K resistor 500 Ohms, the output impedance would be
103K || 500 Ohms = 497 Ohms PRR?

I saw the same Soldano clip Fozzy215. Maybe people don't like the loop because they try to use it with instrument level pedals. I called the design uncommon because I thought that most effects loop sends put the voltage divider before the CF, but I will look up some well known amp designs to check now. The DC coupled CF is undeed something that adds 2nd harmonic distortion, I have a feeling you are already familiar with this link The Valve Wizard

Cool that you spotted the too small cap Gnoboddy, I did not thought about it raising the output impedance, but since Soldano intended it to feed line level gear ( 10K-22K input impedance) I would also make it bigger to make the low end a tad more "transparant".

Does the 1M resistor serve a purpose?
 
So if we would make the 2.2K resistor 500 Ohms, the output impedance would be
103K || 500 Ohms = 497 Ohms PRR?
Plus/minus the usual 10% or more...none of these numbers is exact because of component tolerances. Fortunately, for audio electronics, we rarely care if things are as much as 10% off-spec. (With some exceptions, like high-order active filters.)

Does the 1M resistor serve a purpose?
Yes, it does! It's there to make sure the 1uF coupling cap has a path to ground, so it can finish charging up once the amp is powered on.

If you don't have that 1M resistor, the capacitor won't be able to charge. The left end of the cap is at a slight positive voltage, and when you plug the FX send into some other piece of gear, the input of that device provides a path to ground for the right end of the 1uF cap. So now the capacitor charges through the input of that external FX gear, and you hear a big "Thump!" through your speakers.

So the 1M is there to try to stop "thumps" when you plug the send signal into some external piece of gear.


-Gnobuddy
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.