Looking for EL84 Tone Comparisons by Tube Brand

[pdf64]

...I took out v5 & v7
...
I don't think it will hurt the V33 to use it without v5 & v7...

Why do you think that? It looks to me like V4&6 will idle much hotter. Where do you think they get their bias voltage?
https://el34world.com/charts/Schematics ... 33_212.pdf
Yikes it’s got a switched mode power supply!

[pompeiisneaks]

I don't know I've heard of anyone claiming the tone of their crate was their favorite amp.

Crate Vintage Club was one of my first amps and I have to say it was very good sounding. They have build few very good amps and alot of cheap not-the-best stuff.

This still doesn't negate my point. "very good sounding" isn't the same as "my favorite"

~Phil

[pompeiisneaks]

I don't know I've heard of anyone claiming the tone of their crate was their favorite amp.

Crate Vintage Club was one of my first amps and I have to say it was very good sounding. They have build few very good amps and alot of cheap not-the-best stuff.

Yes - Some Crates have pretty respectable tone, but still get no respect, which is OK by me. A guitarist in a popular Denver band demoed the Crate 32 for me at GC - he played the gain channel with boost on at about 8.5 volume with a Les Paul - The V32 has a pretty great sound like that - a bit of a one trick pony, but you know if the trick is a good one, then it is pretty great. So I bought one. The V33 is really very similar to the V32, but Crate changed it enough to really mess it up. Thing is the V32 would overheat and burn out the input diode (i think that was the component), a really well know problem. I have taken the top grill plate of to give it more air flow, for better cooling - the suggested fix is to get the diode up off the circuit board to prevent the problem, I should maybe do that.

SO ANYWAY just found something out about this Crate V33, been messing with. I took out v5 & v7 - and "OH - I LIKED IT!"

After lots of pre-amp tube changes, I settled on JAN-GE's in v1,v2,v3 and “Reflector 6P14P” in v4 & v6. And the sounds WAS inspiring for me! Which is what I need to play and hear NEW sounds to play better than usual AND to play NEW things I have not played before - just because the sound is inspiring!

I don't think it will hurt the V33 to use it without v5 & v7. But the CAB I used is a 16 ohm where the V33 only has 8 ohm outs. So I need to find out if it is going to be OK to run it like that, because I want more of that!

I think I'll repeat here that you've made it clear you liked it, but nowhere did I see you say it was your favorite amp. It can be very cool/good in a few very specific use cases...

But I am sure I'm wrong and someone, somewhere will say they play nothing but crate because of the amazing tone they love.

[AmpAwe]

yeah NOT my favorite - but inspiring - YES

[AmpAwe]

Why do you think that?

I read it on the web - and did not see a challenge to running with 2 power tubes on the v33 - there was a comment to not pull from the same side, saying that could unbalance the OT and stress it. After playing it for 20 - 30 min I check it for any heat and everything seemed cool, but i did not check inside the chassis. I would have never tried pulling half the tubes, but 4 reading several, of guys doing

mostly - here: https://www.premierguitar.com/articles/ ... _Crate_V33

This is found there - "Moving on to your output stage and speakers, you mention you’ve removed two of the EL84 output tubes so the amp would be running at 18 watts. First, I hope you have removed the correct tubes, which would be one of the tubes from the left pair and one from the right pair. Removing two from only one side would result in no load on one half of the output transformer— a condition I’d advise against.

Also, with a pair of output tubes properly removed from the amp, you have now changed its output impedance. Since you have in essence halved the load on the primary of the output transformer, the secondary of the transformer expects to have half the load on it for everything to be balanced. This means that your output impedance is now theoretically 16 Ω, so I would recommend installing two 8 Ω speakers wired in series for the correct 16 Ω load."

[martin manning]

Well, you can't believe everything you read on the internet. The author of that article failed to notice that the V33's power tubes share a common cathode bias resistor.

[AmpAwe]

Well, you can't believe everything you read on the internet. The author of that article failed to notice that the V33's power tubes share a common cathode bias resistor.

Well Thanks for that revelation – your info led me to read and learn where its cathode resistor is. I am trying to learn to read schematics – that V33 schematics cathode resistor is #R94 is its value 82 ohm? 7 watt. Seems odd to me there is no omega there.

https://www.thetubestore.com/crate-schematics

I looked at the V18 schematic and see it's cathode resistor is #R49 is its value 150 ohm? 7 watt. Close to double of 82 - kind of expected. So then, with only 2 tubes the plate current may be close to double as well. A tube killer, yeah?

I would like to learn how to be able to analyze that cathode resistor circuit and the power tubes voltages and currents, that are in parallel with each other next to the cathode resistor, and to predict the voltage and current expected changes for pulling 2 PT’s but I don’t know how. Maybe I can find an example of that.

Are there modifications that are typical, to allow amps to use half the PT's? Or is it advisable to just not do that?

[martin manning]

The math is simple. Half the number of power tubes requires doubling the cathode resistor to maintain the bias voltage, and twice the load impedance (which you can get by doubling the speaker impedance or, if there is one, moving the impedance selector down one click). To get the correct bias with either two or four output tubes, you could have one pair share a cathode resistor and bypass cap, and the other pair share another.

[roberto]

Those test prove valid and valuable [omissis] to get the amps to sound much better.

This is exactly what R.G. is correctly arguing.

[AmpAwe]

This is exactly what R.G. is correctly arguing.

Yeah OK - point noted. I appreciate you posting a follow up, to repeat the comments, reading them again, I can better appreciate them even more.

I have taken enough math to really come to appreciate how it can very accurately represent and model real world systems. So what R.G. is asking for may be possible, likely is, but certainly not trivial. My mathematics efforts were some years back, mostly now just still marvel at how amazing I think it is, what math can do... They use fractal equations to compress data sets like huge telescope images before transmitting them to earth from space, because the full image might take many of our lifetimes to transmit uncompressed. Then after they get the compressed data to earth they run the fractal equation on the data and poof, the recreated image is restored. So now they use those on any huge data sets for compression to a smaller size. Also the stuff they are doing now... building quantum registers in quantum computers, that, as the theories go, may tap into parallel dimensional realities with the intent of solving "indeterminate problems". Solving an indeterminate problem is maybe what R.G. had in mind... So...it likely is be possible to model "tone" scientifically. But I don't think my brain is up to it, and especially not any more. But math and science is really amazing.

It is really fun to realize these tube amps technology was invented about 100 years ago, and WE STILL LIKE EM!

The retailers and manufactures certainly have some understanding of the useful differences in their tubes. Seems the more they cost the better they is to sound. - and reviews of tubes go on and on about those differences. To wit, some of the expensivenly tubes I have purchased sound to me like they are worth the expense. But OMG are they getting up there.

I never appreciated really expensive wine, till a friend a lot more wealthy than me. Treated me to a couple of glasses of ridiculously priced wine ($2000 bottle was the claim). I could NOT believe how amazing it tasted! Discovering, yeah there IS a reason they get that kind of money for fine wines. Have never even heard talk of putting a science to it - but they probably are talking about that right now, somewhere on a wine forum.

With these tube amps I am just barely yet learning about the circuits.

I DO SO appreciate the helpful information I have received here - think I will look into the suggestion to give this amp the ability to run on different number of pairs of output tubes. May be a fun little project. I do wonder how much of that "tone" that I really liked was from overdriving the OT's having pulled one of the pairs of EL84 - More learning is in order, before tho. First probably need to learn how to measure all the voltages - plates, cathodes, B+ nodes, and –Vbias. And baseline test all those in the amp as it is now before mods.

A far as FAVORITES - I really think of amps like I think of my guitars. Each one sounds different and may better choice for a particular given song. I really like each one for is character. I can admit that I do have favorites - Guitar is Les Paul, Amp is Marshall 2466VM w GoldLionKT66's - is interesting to note this amps gain has 2 separate pots, “detail” for high range, and “body” for the low range. I really like it as a way to control gain. I want to study that set up and apply it to another amp, if I can. Has anyone looked at that?

Thanks again!

[roberto]

With these tube amps I am just barely yet learning about the circuits.

Most of us are doing the same here. Just imagine that still now, after all the decades that passed since the first triodes started singing and after all the decades that passed since the first vacuum tubes became obsolete, people are still inventing new ways to improve their working conditions. Very few people worldwide can say to master the argument.

[roberto]

Near as I can tell, any time you nail down the facts, the magic disappears.

Question of scalar tone aside, is there any research on a decent number of tubes of different producers (not brands, but producers)? EG, what I've noticed is that JJ 12AX7 are usually more sensitive to grid stoppers (higher Cga?). This is something I cannot measure, but based on many comments also from other people, it seems a common feeling they give. Is there any analysis like that, fully available online?

[R.G.]

Sadly, probably not. Or at least, not in any useful form. What follows are my guesses about the situation, based on 40+ years of musical electronics and an EE career.
I think musicians are most interested in things that are not on datasheets. Triodes without feedback are very linear amplifying devices, but they are not perfect. The distortions are small, but do exist. I think that what distortions are there are, as you were thinking, based on subtle differences of the physical structure of the cathode, grid, and plate and how the ballistic flight of electrons from the cathode is on average messed with by the charges on the grid. The input and feedback capacitances are probably part of it, as you've guessed, and probably nonlinear to boot.
Triode distortions are small, but there. I think that you could get some useful information by using today's very, very sensitive audio analysis instruments and mapping the distortion of a tube for various sizes of grid signal over the operating range of the tube. This would have been practically impossible in the days when you could not yet tell a computerized instrument "now go run all that again on the next unit".
Electronic components exist as datasheets, not physical objects so much. Any component you can buy is an individual. But to be labeled and sold as a "12AX7" it has to be within the minimum and maximum ranges of the component datasheet. Economics leads manufacturers to be persnickety about this. They craft the datasheet to (1) sell as many of that kind of device as they can and (2) specify the widest limits they can get away with so they throw away the fewest rejects that don't meet the specs. So they certainly don't specify things like sensitivity to grid stoppers and mid-band distortion over signal ranges. And if it's not specified, it's certainly not looked at by the maker, and will >>actively<< not be considered in when changing or improving the manufacturing process.
That means that the manufacturer brand (which gets sold and traded around a lot) and part number/datasheet cannot be relied on for any kind of perceptual insight on a tube. The manufacturers change, the machines that make tubes change and get readjusted, and the materials drift over time. So even if you could measure the human perceptual response to some given tube, the manufacturer and brand of any given tube is not a reliable pointer to whether buying another one of that same manufacturer and type number will sound as gratifying.
The single most likely thing that will affect whether you will think a new >>anything<< is good and satisfying is whether you believe it will be. Your expectations shape your brain's perception. In extreme cases, this leads to the placebo effect. Your brain just makes it up, below the level of consciousness. In subtle cases, you get slightly perturbed impressions. The hifi tweako bunch knows this, after blind testing invalidated the opinions of Golden Ear audio critics over and over and over. It was so bad that the tweakos now have retreated to simply not cooperating with blind tests, and to rationalizing that the simple act of testing makes the magic disappear.
Well of course it does. That >>is<< the point. Your brain is much happier with what it actually hears, plus the generalized goodness it expects. Take away the expectation, and it's less satisfying. Most people don't like numbers.
I think some programmed and sensitive automated testing of triodes for mid-band distortion and such could turn up some patterns of good distortion and bad distortion. We pretty much know it's in the distortion, because blindingly linear and distortion free sound devices get criticized for being "sterile sounding", right? So it is at least theoretically measurable. But it would take a big, well instrumented effort to measure it. One guy with an oscilloscope isn't likely to get the wide sweep of info that's likely needed.

Just sayin' ...

[roberto]

Thanks for the deep explanation and for opening another window in how a well implemented Arduino-based equipment could add some more light to the knowledge of tubes.
Is it anything you have already planned to design/build?

[R.G.]

Sadly, no. that polemic was an off the cuff synthesis of previous thoughts that the thread brought together.

If I was at the other end of an electronic/music career, I'd probably pursue it. But the actual design is more pick and shovel work from here on. There are some programmable audio analysis machines that already exist. It might be as simple as either writing some macros for a possibly-existing Audio Precision instrument, or constructing a program driving interface plus some outboard programmable power supplies and test sockets.

The human and economic side of the issue would probably weight against it. It's likely that neither the tube enthusiasts nor the tube makers would really like to see black and white columns of numbers. That takes all the human-side fun of arguing about whether one tube is provably better than another out of it. And it would most likely not change anyone's opinion. They'd simply attack the data gathering as flawed.

Oh, wait. My attitude about the variability of humans is showing, isn't it?