I am brand new to the world of tube amps. I have a '55 AMi G-120 Jukebox with a mono 6L6 amp that a knowledgble friend helped me rebuild. I've been bugging him about modifying the amp to clean up the sound and improve the headroom with 6550's. He's an excellent engineer, but not good at explaining why I can't just drop in 6550's and make 1 billion gigawatts. This video showed very clearly the pitfalls of tuning an amp "by ear". Best 15 minutes I've ever spent on YT. Thank you 4 posting!
That is correct about the 7027 - pin 1 is tied to pin 4 internally which is the screen - which also means depending on how the socket is wired, if a 6550 with its metal ring is plugged into a 7027 socket you could end up with screen voltage on the metal ring (because of pin 1). Also, pin 5 and 6 are tied together on the 7027 which means on the 6L6, etc group of tubes, often times pin 6 is used as a tie point. Thanks for the post and bringing this to light for all viewers.
Man... I have to admit I know very little about all of this... Some but not much but watching these meters display what's going on is so facinating. I can't wait till the day it all makes sense thanks to all the TH-camrs that spend the time to show us this stuff.
this is so cool. I am a worship pastor and I use tube amps with my electric gutiars and I love how all this works. I hope to have the understanding and some of the testing equipment you have someday. Glad that there are people doing this kind of stuff still, no one my age seems to be (Im in my mid 20's). The only tube thing Ive built so far is a 12ax7 powered overdrive pedal.
Yes, you should adjust the cathode currents when you change tubes. You do this by measuring the voltage across a cathode resistor (it it has one for this purpose). An example is the Fender DeVille which has a 1 ohm resistor from the 6L6 cathodes to ground and a designated test point which you adjust to 60 mV. This gives 60 mA current for the two tubes. If only a bias voltage point is provided or you measure directly at the grid, it shouldn't have change when tubes were replaced. Hope this helps.
Bias adjustments are meant to set a value of cathode current which is definitely related to plate voltage. The cathode current is the sum of all the currents in the tube (i.e. plate and screen) but the screen current is small so essentially, the cathode and plate current are the same. The Current is what you actually want to set and ultimately you want the tubes to be "matched" by having the same current (measured in mA) flowing in each cathode of the output tubes which makes them "balanced".
I have noticed that the vintage amps deliver a lot of power with low distortion and no clipping up to full power - example - Ampeg SVT. And I agree with you on the Fenders (6L6) - usually run at a lower distortion (overdrive) level and cleaner tone and the Marshall's (EL34) run at high power and lots of overdrive/crunch (which is clipping at 10% THD or better) They all have their purpose in making the music we love.
I completely agree with you that output transformer impedance is critically important. No doubt this plays into this demonstration. The primary goal of this video was to give some empirical evidence of (some of) what is actually going on when output tubes are swapped with another type. The higher bias voltage on the 6550 (may) allow the smaller tubes to be swapped in without immediate destruction but certainly not recommended for long term use. Thanks for your comments.
Remember that power is voltage squared divided by resistance - so, for an example, 20 volts across 8 ohms is 20 squared (400) divided by 8 which is 50 watts. For 50 watts across 4 ohms it would be only 14.14 volts. Across 2 ohms it would be 10 volts for 50 watts. Hope this helps.
Yes, 420 - 450 is a healthy plate voltage for 6L6's although 490 or so is not uncommon and I believe that is what the Dynaco and MA230 cloan is running at in this video.
Great Video, you must be an O'l TV repair man with all the knowledge you have, thanks for taking the time to educate some of the students of electronics out there., and please keep up the good work, I myself is can't thank you enough for the free teachings that I'm sure you had to pay for through out life. Thanks again and people out there if you don't like the video don't watch it,, it's that simple.
EXCELLENT. Once again a sober, well researched test based on science, repeatable measurement and not some air-head's "killer ears". All it takes is some reading and LEARNING from text books. The ARRL handbook is a great primer. The math is simple, but verbal hype from the guy at the bar or the salesman at the snobby hi-fi shop will NOT teach you about this technology. Most people hate having to learn even simple science. Lazy brains I suspect.
You got that right. I sold my '66 Marshall plexi 50w years ago because if I turned it up to where it sounded good, everybody in the world hated my guts. These days, I use an old Tweed Fender Princeton and everybody's happy, including me. That's it in the little picture to your left. It works great.
I think over all, that is good advice from Dr Z for most people but I have never been one to just take things at other people's word. I need to understand and find out for myself. I am now 62 and have been this way since I was a small child. I like to know and understand from my own measurements. This attitude carries with it some risk and cost but also the possibility of much real gain. Research your subject well but don't be so timid as to never be able to Plug-It-In and Turn-It-On to learn.
With correct cathode current of 40 mA on the 6L6 raised to 80-100 mA, I would say less than an hour. I used to have a McIntosh MC240 that would ruin 5881's within the day. 6L6GC would last years. If you look up the old specs of tubes you will see that a 5881 has less plate dissipation than the 6L6GC - and plate dissipation matters very much. If you can measure cathode current, just simply multiply that times plate voltage and you will get idling plate dissipation. It will increase under drive.
PS: I'd love to see a proper show-down of the 6L6 vs. EL 34 (for hi-fi purposes). Do a sweep through all extreme freqs, IM, HD, clipping, headroom, periodic harmonics, etc. E.g. do the 6L6s really have better bass? Which valve has better transients? @ which freqs? Using the same amp (output & power x-formers) with reasonable bias, cathode current & B+ adjustments makes this a 1st class objective test, Next, I'd love to see tube brands tested with no tears. -GREAT STUFF!
The Fender Bassman 300 I have a video on has around 750 volts on the plates of the 6550's. Certainly no place to plug in a 6L6 or 6CA7. 450 volts on the plates of the 6L6 is not uncommon and is typical in the old McIntosh amps. I agree with you that for a typical 6L6 style amp, 6.5Kohm output transformer performs well - much better than a 5Kohm. Thanks again.
Your video is really good! I learnt many things about tube replacement, and the idea of the BIAS pot on top of the chassis is interesting. I'll try all this in my amps, Thank you ;)
jibber jabber volts amps jibber. I wish i had the background to understand what you're doing. looks like really cool stuff. I used to see mcintosh amps in pawn shops 25 yrs ago. I wish i grabbed a few when they were there. Oh well. I got here while watching vids on adjusting/modifying tube guitar amps
Fortunately my wife has stayed with me 35 years, so far. The first one didn't though :-) As for 3 main types of distortion I assume you mean amplitude, harmonic, and frequency? And the old amps often times specified phase distortion. I will have to think about it and see if I can come up with something real. Thanks for your comments.
Pin 1 and pin 6 on the 6L6 is not used and often times used as solder connections, on the tube socket, for component connections under the chassis. The 6CA7 pin 1 is the screen grid and is most always grounded or connected to the cathode. The screen grid is internally tied to the cathode in the 6L6. So, you MUST know how the socket is wired up before swapping tubes around. A socket can be wired to take either of these tubes. Check out the pin configurations before swapping tubes. see next post
With all said below, if you can measure cathode current and limit it to about 140 mA with no drive for the pair of KT88's then it might be fun to try. EL34's draw 1.5 amps per tube (3 amps total for two) so, if your amp will support EL34's it should do OK with the filament supply for the KT88's. Reasonable cathode currents for a pair of 6L6's or EL34's is about 80 mA (40 mA per tube). Hope this helps.
I have never personally had any KT77's but they are sold as replacements for EL34. The KT66 is sold as a replacement for the 6L6. KT88's and 6550's are interchangeable. All, of course, with attention to bias and plate current when initially setup.
I can't remember the tube layout for every amp so I make mistakes. I have done mods for clients whereas I have swapped out 6L6's for KT88 and the client is happy just as you are. The potential problem is the 6L6's draw 0.9 amps each for the filaments and the KT88's draw 1.6 amps each so, most likely the filament voltage is low and the power transformer filament winding is overly stressed. It may last years or it may not. If it works and you are happy with it then all is good. Not all amps (next)
I understand what you are saying but I came to realize that the creation of music and the playback of music is two different worlds and the fact that tubes are electro-mechanical devices and are affected by the vibrations of the cabinets is what allows musicians to create the music we love. Then, we try to playback at ultra-low distortion the music they created a 10% THD and above. Holding the guitar in front of the speakers may create a desirable tube related mechanical feedback that we like.
Best to tube it for what it was designed for. KT88 and EL34 draw about the same amount of filament current (1.6A and 1.5A respectfully) 6L6 is only 0.9A so if you put the bigger tubes in a 6L6 amp, you will overload the filament winding of your power transformer. 6L6 and EL34 are close in power output. You can get a little more out of the EL34 as it can take 800 volts on the plate, 500 max for the 6L6. KT88/6550 is twice the power tube of a 6L6.
In a nutshell, you have to re bias your amp if you change the tubes. Some amps have switching on the back, others the bias has to be manually done. So, the rule of thumb is just use the recommended tubes. My experience is the better the tubes the better your amp will sound. New tubes have to be broken into, they will sound very clean. I was upset when I put new tubes into my amp, it didn't have the same distortion and sounded too clean. Well, a lot to be said about old tubes and good distortion.
So, to make the socket ready to accept either the 6L6/KT88/6CA7 - you Must remove anything soldered to pin 1 of the socket and either ground pin 1 or connect it to pin 8 (the cathode) - probably best to ground it - Connecting pin 1 and pin 8 together carries a certain risk as pin 1 on metal based tubes have the metal ring connected to pin 1 so you could end up with the bias voltage on the metal ring of the tube and it will bite you if you touch it with the amp powered up. see next post
20/20 hindsight. I can only hope my retirement funds start moving soon because, according to their growth the past 2 years, i've been in some sort of time stopping bubble where nothing gains value. I also remember nakamichi dragon tape decks in the pawn shops and i was drooling all over those too, given the choice, i could see myself getting that instead of the power amps because i was looking at all kinds of amps. There was so much to choose from compared to top shelf tape decks.
6L6's run at about 40 mA each (80 mA per pair) and it is customary to put in a fuse at twice the normal current rating. KT88's are about 70 mA per tube (140 mA per pair) so,l with that approach you would need 280 mA fuses - It would be wise to stick with the 6L6's - there are many brands out there, and 6L6's are about half price of KT88's. Hope this helps.
@MICKSHRED No, that would be a bad idea. As documented in one of my later videos, filament current of the 6L6 is 0.9A and the 6550/KT88 is 1.6A so you will overload your filament winding in the power transformer. Also, the 6L6 runs at about -40 volts whereas the KT88 runs at about -46 volts so the KT88's would be drawing much more cathode current (once again) overloading your power transformer. The other way around (might) be more acceptable.
Pin 1 on the metal based 6550's is connected to the metal base and should be grounded for safety. Pin 1 on EL34's is the third grid and must be connected to ground or the cathode (pin 8). G2 at 1Kohm sounds like a smaller tube than the 6550. G1 at 55Kohm is pretty high. Glowing hot EL34's - not good - make sure the bias voltage is adjustable to at least -60 and there is no AC ripple on it. My guess at a distance is bias supply issues. Hope this helps.
The 6V6 is a beautiful low power tube and it only requires 0.45 amps filament current (you might think of a 6V6 as 1/2 the power of the 6L6. The 6L6 requires 0.9 amps filament current. The EL34/6CA7 requires 1.5 amps filament current so it would seriously overload the filament winding of your 6V6 power transformer and I doubt the 6CA7 would even light up properly. So, unfortunately, no, this won't work.
Well, EL34's draw almost twice as much filament current as a 6L6 so the filament windings won't be overloaded and the bias is "probably" high enough (negative enough) on the EL34's so the 6L6's won't draw too much plate current. However, the EL34's can handle much higher plate voltages (800V) than the 6L6 (500V). Overall, it might work "OK" but probably best to put the EL34's back in. Google "6L6 datasheet" and "EL34 datasheet" and check it out for yourself (nice PDF from Svetlana out there).
6L6's are really good to 40 watts and 50 watts max with any reasonable amount of distortion in the creation of music. If you drive it into a square wave output you may get 60 watts. Similarly, the 6550/KT88 is outstanding at 75 watts but will do 100 watts at higher distortion. Square wave output on a KT88 might reach 120 watts. Just because someone puts 160 MPH on the speedometer of a car does not mean it will go that fast - unless it is off a cliff and it won’t be long before the ride is over.
If you check plate voltage with no tubes the power supply will be operating with no load and it will be abnormally high and not the voltage that will be on the plates of the tubes when they are plugged in. The bias voltage will not be affected with or without the tubes being in.
I think it's a personal preference and each tube type seems to have its own charming characteristics but overall, large and small, I am partial to the sound of the EL34/6CA7. Thanks
With an el34, you need to plan on jumpering pin#s 1 and 8 in order to get the suppressor grid polarized. Most guitar amps use open pins on the octal socket to support grid stopper and screen resistors. Since the 6l6 cathode is connected internally to the suppressor grid, no problem! Plug in a el34 into a 6l6 amp blindly might not be a terrific idea.
I think we all know about pin 6 as you mentioned, to mount a grid stopper resistor. As for pin 1, it should not be used for a tie point (I learned this the hard way) as it is connected to the metal rings of tubes like the KT88. I once used pin 1 for a screen resistor and you know what happened when I accidently touched the metal ring. So, whether it is a 6L6 or EL34 it is a good idea just to go ahead and ground pin 1. Leaving the suppressor grid open on the EL34 will probably just keep the tube from conducting very well and power will be low and distortion high. I haven't verified that but I think is what is likely to happen. Thanks for your comments.
I don`t know if you are using grid or cathode bias but check the manufacturers tube spec. Also remember that with EL34`s (6CA7) you must earth the screen pin (G3) seperately.
The extra power sure helps with bass and transients. Frequencies below 200Hz use most of your current. You speaker is 8 Ohms @ 1000Hz (NOMINAL impedance). At 80 Hz where the deep bass of the kick drum or string bass resides, most speakers go down to 1-2 Ohms -or less! -That's where all that great KT88 current goes! Now they even make a KT120. But if you get up to that power range, Why not experiement with 811s or 572Bs (they're not expensive) -You might like class A at 50 watts!
I wish I had bought every Mac amp I saw even at retail price back in the early 60's - the MC275 was $444.00 new - a NIB one now from that era would be worth an easy $10K+ - better than the stock market. I love this stuff and now that I am retired I get to play with it a lot - I don't have any vids on mods - I have done mods for musicians around my area but it seems many turn out disappointing as they don't get the sound they thought they were going to get - good luck.
Excellent presententation. You, um, have a few more test toys than I do. You are either a serious pro or your wife has left you. Seriously, This shows that there is not much difference in the desgin for push-pull amps using the popular valves. However, it also shows that there is a more than simple grid bias to good tube life, low distortion & high efficiency. Could you do a test on power vs. 3 main types of distortion with the 4-5 main power tubes?
You should be able to plug your KT88's into your 6CA7 amp (if and only if) you have enough range in your bias adjustment to lower the cathode currents in the KT88's to 70 mA per tube. The socket as wired for the 6CA7 should be OK for the KT88 and as I mentioned below, the worst case would be bias voltage on the metal ring of the KT88's. Check your socket. If pin one is grounded there will be no bias voltage on the metal ring. Work safely with High Voltage.
Yes, I have always heard that the 6L6 is a better bass amp tube than the EL34. Not sure how to measure that. Again, got to think about it. Thanks for your comments.
It might need new tubes but It is wise to stay with the same output tubes. Your amp is designed for the EL34. ECC83/12AX7's can last a very long time as they run much cooler. Adjusting bias - that's a whole other discussion. Using a good quality contact spray cleaner does miracles and may be all you need. Dirt, grit (and beer) take a serious toll on musical amps. Spray/clean every pot/switch/and plug if you are comfortable with working on your own amp. High voltage in tube amps can be LETHAL.
Those 6CA7s tubes are very dependable and they sound great I have a Quid set in my plush amp we tinkered with they are GEs however so the tone is there. EL34s were very unstable in the plush. I did put in a presence pot in the circuit. so the amp sound like a fender on steroids a cross between English & American. I like the math of it but I am not a tech. Thank you for posting this video. c];-D
And lastly, remember that these "experiments" in swapping tubes around are not meant to be permament changes. KT88's are more powerful tubes than the 6L6 or 6CA7 and will require more energy to be delivered by the power supply. Ultimately, my goal was to answer a client's question and my last and final recommendation is to Not swap around your tubes unless you are knowledgable and experienced with working with vacuum tube amps and are willing to accept the risks to yourself and your equipment.
Thank you for your comments. There are may reasons we can't permamently swap 6550's into a 6L6 amp and one of the most fundamental reasons we can't is because of the filament current required for the 6550's. A pair of 6550's require 3.2 amps to light the filament whereas a pair of 6L6's only require 1.8 amps. The filament winding on the power transformer will be damaged over a period of time if it is overloaded by that amount.
today, all i can say is thank goodness i did not buy one because i i would need to go buy cassette tapes in the clearance shops beside VHS/Beta tape display, which is behind the film camera aisle, around the corner from the wooden wagon wheels. I still like my vinyl but it's pretty much impossible to find new stuff these days and besides that, the analog edge is gone now that mastering is pretty much all digital so tapes and vinyl makes less sense to me every day. I enjoyed the video.thanks
The 6CA7 filament draws 1.5 A and the KT88 draws 1.6 A. This probably OK as it only 200 mA difference and (probably) will not harm the power transformer. The 6CA7 and KT88 can both take 800 volts on the plate whereas the 6L6 can not. Also, the 6L6 has a filament current of only 0.9 A so putting 6CA7 and KT88 in a 6L6 amp will probably grossly exceed the filament current rating of the power transformer. see next post
I will have to think about this one. It will be a challenge. Some of my first videos are comparing KT88/6550's from vintage to modern Chinese and Russian in a Dynaco amp. You might find some of them interesting.
A pair of 6L6's is good for 50 watts and a pair of KT88's are good to 100 watts - that's just the way it is. You cannot put KT88/6550 in place of 6L6's - It will "work" but not very well and not very long. Hope this helps.
No fair! You're using a spaceship navigation computer! (If you ever see 'Silent Running,' look for the Tektronic oscilloscope on the bridge of the ship, Valley Forge.)
KT88's draw 1.6 amps per tube filament current to a 6L6 of only 0.9 amps per tube so, that is one concern. The bias on the KT88 will need to be made considerable more negative to reduce the cathode current to an acceptable value. There is likely not enough adjustment in the bias pot to accomplish this. You will need to change bias divider resistors and even then, you may not be able to attain a proper negative voltage setting for your KT88's. You may find more help in some of my other videos.
I read an article on the KT88 tubes. Swapping from EL34 to KT88. It read that the bias feed resistors should be switched from 220k to 100k resistors to accomplish a proper reading. I believe it was something about reducing the grid resistance, so the KT88's can run at the proper rating. Is this true? and where are these bias feed resistors located?
Speakers being connected is only important if you are driving the amp with a signal. You can safely measure voltages with no speakers attached. High voltage is a rattle snake so connect your leads before you plug it in to the wall. Work with one hand in your pocket and never lean on the equipment and don't use both hands when measuring high voltage as you want to avoid any current flowing from arm to arm through your heart. Getting across the plate voltage is a bad experience :-(
This seems to be a very informative video -- thank you for posting it! In the spirit of improving things, I must say that I did find it difficult to watch because of the head-mounted camera, and had to stop watching part-way through. Perhaps a tripod would help, and as would arranging your test equipment so that the DVM and scope are both near the amp can be seen with minimal camera movement. Not trying to be mean-spirited, just a suggestion for your consideration that might get you more viewers.
Hi there! Very interesting video! I am new to tube amps and just got a Peavey 5150. Eurotubes says that if you install an integrated quad set of 2 JJ KT88's or 2 JJ KT66's and 2 JJ 6L6's then the amp does not need to be biased and it's basically "plug and play". They said they did a lot of research on this and that it not only sounds great but saves you from having to bias the amp. They recommend you put the 2 KT's on the outside sockets and the 2 JJ 6L6's in the 2 middle sockets. You seem to know what you're talking about so I was just wondering if you think this is safe and is a good idea. I know this is an old video so I hope to get a response cause I just ordered the tubes today! Thanks!
I have only experienced the KT120 in some Dynaco spin-offs - can't remember the brand - impressive tube, very healthy bottle. I think their cathode current is a bit higher than the KT88 and KT90 so. some care should be taken before switching them.
Even though It's done everyday, I was always afraid to put EL34's in a 6L6/6550/KTxx/ socket due to a slightly different pinout. 6L6 and the like use a common (7AC) socket where pin 1 is not connected (NC) and pin 8 is connected to the cathode (k) and grid 3 (g3) where as the EL34/6CA7 (8ET) socket uses pin 1 connected to grid 3 (g3) and pin 8 is connected to the cathode (k) on its own.
Hello. Just catching this video for the 1st time. Great video. You keep referring as 6L6 but arent those 6L6GC? Theres a big difference between to the 2.
Is that where the stainless McIntosh (repos) are coming from? I haven't seen them on Ebay yet but I have seen the "redone/rechassis" amps that look great. If someone starts making repo Mc transformers I have serious concerns that they will be the quality of a Mc transformer which is the heart and soul of the Mc performance. Thank for bring this up.
Now there is a handy and safe tube puller / inserter available on ebay, item 254242318237 Easy, simple and safe. A perfect tool for any audiophile or vintage electronics technician.
Thanks for noticing and correcting. I never noticed and now that TH-cam has taken away the "bubble" type comment that I used to be able to make and correct mistakes I can no longer make changes to my videos.
I am not familiar with your amp so I can't say for sure. 6L6 & KT66 are interchangeable as is the EL34 & KT77 but there is considerable difference between 6L6 & El34 and again, a considerable difference with the KT88/6550 - basically, a KT88 is capable of twice the output power of a 6L6. EL34's have a much higher plate voltage rating than the 6L6. Without a considerable bias voltage range the three tubes above can not be substituted one for the other. There is also filament current issues.
I don't have my tube specks book but the diference on these tubes are the plate resistance I know that you could get up to 70 w from a pair of 6550's and only 50 w from a pair of 6L6's of coure adjusting the bias for AB1 mode but if the output transformer is not matched to the plate resistance that I believe is 6.5 Kohms for the 6L6 you are not going to have optimum efficiency and I don't remember which amp but it had 6L6 with 475 volt plate voltage most run the 6L6 at 350 to 400 plate volts.
I too am contemplating a KT88/6550 tube change in a older marshall 1988 for more clean headroom, none of the new amps have the tone i’m after and I can’t afford a Dumble or Bludotone!
+R. Weaver It could be OK and you may like the change in tone. If the Marshall runs EL34's the filament current change to the 6550 will be OK. The filament of the 6550/KT88 and EL34 draw about 1.5 amps each whereas the 6L6 only requires 0.9 amp each) What I hope was noticed in the video is that the bias voltage for a 6L6 is approx -37 volts, for a EL34 approx -44 volts and for a 6550/KT88 approx -55 volts ( all approximately). What the point is, if your Marshall will allow you to adjust the bias a bit more negative, the 6550 could be OK but you must check it or you will just ruin tubes. Another point is, with a 6550 amp with a bias of approx -55 volts, even with the higher plate voltage in the EL34/6550 amp, you can, for a short time, put a 6L6 in the socket and it won't go up in flames since the 6L6 typically needs only -37 volts bias to set the cathode current at a safe level. But, in a 6L6 amp with -37 volts negative bias if a 6550/KT88 is placed in the sockets, the tube will draw Way too much plate/cathode current and things will not go well. Also, placing either EL34 or 6550 in a 6L6 amp will overload the filament circuit of the power transformer. Good luck with your project.
ElPaso TubeAmps Tranny is shorted at present, so changing to accommodate tube change should help. I am currently looking to change mid control to enhance 500k a little more.
soo?? this second amp has a B+ of around 350v when the 88s load it down...in order to qualify 70ma each. i am looking to convert a marshall superlead (34s) to 88s.. could you verify the needed change in splitter resistors and screen resistors...?
I would like to try EL34s in my Scott 296. Can it be done, replacing GE7581A(KT66) with 6CA7/EL34s? The 7581 is similar to 6L6GC with more output 35w vs 30w respectively.
May I ask why you switched amplifiers in the midst of what would have been a great comparisom. Unless the B+. screen, OT, are exactly the same there is no ability to compare. Additionally the 6L6s were not equal as indicated by the difference in getter oxidation.
question, when your figuring out the wattage output of an amp, i know you use voltage squared divided by the ohms gives you the wattage, but when your checking the voltage do you have to have to use a true rms meter, or is there a way to do it with a regular multi meter?also, do you know of any cheap devices that I can get to verify the actual output of an amp before clipping? or do you basically have to have a osciloscope, frequency driver, rms volt meter, resistive load(to emulate speaker), and distortion analyzer? or can i not use an osciliscope and just get away using a distortion analyzer?(somehow i have a feeling a dist analyzer is more expensive then a scope) sorry for my igrorance but i'm relatively new at electronics repair and am trying to set myself up for working on tube/ss power amps.
You have described pretty much the whole setup. A distortion analyzer and an audio spectrum analyzer is probably the most expensive things to have. An oscilloscope, a signal generator, a RMS reading volt meter, and dummy loads are necessary. You can drive the amp up to just below clipping and get a pretty good idea of its performance with a THD around 1%. Noticing the shape of the sine wave is important also. Pretty much any volt-ohm meter will work for reading RMS voltage at 1 KHz like the vintage Triplett and Simpson analog meters Knowing the frequency response of the meter before purchasing is a good idea and ideally you would want one that is flat from at least 20 to 20 KHz or better. The Tektronix TM500 series instruments will let you have an oscillator, voltmeter, distortion analyzer all in one mainframe. The most expensive items for this type of setup would be a SG505 oscillator and the AA501 distortion analyzer - you could pay $1K for those items but possibly less. You can buy good used Tektronix scopes for little money and in my opinion they are definitely the best to have. Of course new ones with a spectrum analyzer and oscilloscope built together (called Mixed Domain Oscilloscopes - MDO) are lots of thousands of $$$. Good luck with your enterprise. If there is anyway I can help, please let me know.
have a bias supply with enough range to be able to make the bias supply negative enough to move from 6L6's to KT88's. 6L6's need bias something in the -40 volt range whereas the KT88's need bias voltage in the -50 volt range. I don't know about your Mesa Boogie Triple Rectifier - some Mesa Boogie's don't have adjustable bias voltage - you will just have to investigate the schematic carefully to determine if it will work and not overheat.
hello sir this video is really teaching a lot and it does help me a lot I still reading and comprehending more about tube amplifiers, I have a question if you don't mind, I still in the process of making a KT88 push pull ultra linear hi fi amp "not for guitar" home use on the cathode of the KT88 what resistor should I use 10R 3W is fine or should I get a 5W ? many people have use 5W and 3W what is correct ? the B+ will be around 460V DC from a Edcor transformer model XPWR247-120 330 - 0 - 330 you mention already "long time ago" that I should use 4.2K output transformer one more question about output transformer, should I get one that can go up to 100W or one that supply about 60W ? I was thinking about the B+ and I'm not sure what transformer should I get because of the 460V DC B+ I'm a bit confuse to be honest Regards Juan
Hi Juan - here is the way I would look at it - absolute maximum cathode current for a pair of KT88's is about 275 mA - that is the current if the tube is being driven to maximum sustained sine wave power. (from a 1962 RCA tube manual) That current flowing through the 10 ohm cathode resistor would dissipate power of (I^2*R) which would be about 0.76 watts so a three watt resistor would be OK - even a 1 watt resistor would not burn up but would get pretty hot. Another possible way to look at it is, in case something goes terribly wrong in the tube and there is an excessive amount of current flowing through the tube, a 3 watt resistor would burn up before a 5 watt resistor would and "might" help protect the output transformer from excessive over current. So, we can see that making everything larger is not necessarily a good way to go. On the other hand, if the primary of the power transformer is fused properly, then it should blow and protect the amplifier. Another way to protect the circuitry is with a fuse in the high voltage line - a 0.5 A should be sufficient. Many guitar amplifiers have a HT fuse. Once you complete the amp, if you have one of the handy "Kill-A-Watt" meters that you can get pretty cheap at Lowes' you can actually measure the power up current of the amp and set the line fuse to some value above that point, somewhere around twice the max current that you measure. Hope this helps.
ElPaso TubeAmps good idea about the HT fuse and the cathode resistor to be 3W and yes this does helps about the "Kill-A-Watt" really good idea I didn't think about that, I'm going slow with the build so when I have something done I will let you know thanks for the help :D and yes I have been aware of high B+ dangers don't worried I be really careful my MOS was 94F Special Electronics Device Repair you have a great day sir :D Regards Juan
Can i Replace in a Marshall 1986 or 1987 the EL34 with the KT88-94 for More clean Headroom ? The Heater current of the KT88 is about the same of the EL34 ....I know i have to lower the bias feed resistor from 220k to 100k to avoid tube problems ..the question is : the transformer can handle the bias current draw with the KT88... i have read that the tube should be biased like a EL34...but if bias this way they should be biased too cold !!!!! Can i bias at 60% with a plate voltage of 470v-480v without overheating the transformer ?
+Giuseppe Loriga You are right about the filament current being close in the EL34 and 6550/KT88 so that should be OK. Plate voltage on both tubes can be as high as 800 volts. I doubt it will damage your transformers but I would limit the plate current in each KT88 to 70 mA (140 mA for both KT88's) by adjusting the grid bias. As the amplifier is built for an EL34 the bias is probably not negative enough for the KT88 so it will have to be adjustable to a more negative voltage. Pin 1 of the EL34 is g3 and is probably grounded at the socket. Pin 1 of the KT88 is attached to the metal ring of the 6550/KT88 - make Sure pin 1 of the KT88 socket is grounded to the chassis - as it probably already is..
+ElPaso TubeAmps Thanks for your Help ... So what is the effect of the reduced plate current in the tube ? Can i bias the tube at 60% ( 43.8ma with a plate volge of 480v Class AB Fixed Bias) or it will run colder ? Last question the pin1 in my el34 tube socket is grounded to the chassis through the 1 Ohm resistor , is that ok for the KT88 ?
It's all about the pre amp section, also I don't get why Guitar players insist having the Amp head on top of the speaker stack. Man all the 12AX7's will go microphonic so fast compared to a Audio system that isn't going through a Earth Quake of vibration.. Kills me seeing nice OLD Tubes going into Guitar Amps sitting on the stacks!
I am brand new to the world of tube amps. I have a '55 AMi G-120 Jukebox with a mono 6L6 amp that a knowledgble friend helped me rebuild. I've been bugging him about modifying the amp to clean up the sound and improve the headroom with 6550's. He's an excellent engineer, but not good at explaining why I can't just drop in 6550's and make 1 billion gigawatts. This video showed very clearly the pitfalls of tuning an amp "by ear". Best 15 minutes I've ever spent on YT. Thank you 4 posting!
That is correct about the 7027 - pin 1 is tied to pin 4 internally which is the screen - which also means depending on how the socket is wired, if a 6550 with its metal ring is plugged into a 7027 socket you could end up with screen voltage on the metal ring (because of pin 1). Also, pin 5 and 6 are tied together on the 7027 which means on the 6L6, etc group of tubes, often times pin 6 is used as a tie point. Thanks for the post and bringing this to light for all viewers.
Man... I have to admit I know very little about all of this... Some but not much but watching these meters display what's going on is so facinating. I can't wait till the day it all makes sense thanks to all the TH-camrs that spend the time to show us this stuff.
this is so cool. I am a worship pastor and I use tube amps with my electric gutiars and I love how all this works. I hope to have the understanding and some of the testing equipment you have someday. Glad that there are people doing this kind of stuff still, no one my age seems to be (Im in my mid 20's). The only tube thing Ive built so far is a 12ax7 powered overdrive pedal.
One of the more interesting videos I've seen in a long time. Excellent.
Yes, you should adjust the cathode currents when you change tubes. You do this by measuring the voltage across a cathode resistor (it it has one for this purpose). An example is the Fender DeVille which has a 1 ohm resistor from the 6L6 cathodes to ground and a designated test point which you adjust to 60 mV. This gives 60 mA current for the two tubes. If only a bias voltage point is provided or you measure directly at the grid, it shouldn't have change when tubes were replaced. Hope this helps.
Bias adjustments are meant to set a value of cathode current which is definitely related to plate voltage. The cathode current is the sum of all the currents in the tube (i.e. plate and screen) but the screen current is small so essentially, the cathode and plate current are the same. The Current is what you actually want to set and ultimately you want the tubes to be "matched" by having the same current (measured in mA) flowing in each cathode of the output tubes which makes them "balanced".
I have noticed that the vintage amps deliver a lot of power with low distortion and no clipping up to full power - example - Ampeg SVT. And I agree with you on the Fenders (6L6) - usually run at a lower distortion (overdrive) level and cleaner tone and the Marshall's (EL34) run at high power and lots of overdrive/crunch (which is clipping at 10% THD or better) They all have their purpose in making the music we love.
Ive been a big fan of KT88's over the years just something about the tone that i love so much
I completely agree with you that output transformer impedance is critically important. No doubt this plays into this demonstration. The primary goal of this video was to give some empirical evidence of (some of) what is actually going on when output tubes are swapped with another type. The higher bias voltage on the 6550 (may) allow the smaller tubes to be swapped in without immediate destruction but certainly not recommended for long term use. Thanks for your comments.
Remember that power is voltage squared divided by resistance - so, for an example, 20 volts across 8 ohms is 20 squared (400) divided by 8 which is 50 watts. For 50 watts across 4 ohms it would be only 14.14 volts. Across 2 ohms it would be 10 volts for 50 watts. Hope this helps.
Yes, 420 - 450 is a healthy plate voltage for 6L6's although 490 or so is not uncommon and I believe that is what the Dynaco and MA230 cloan is running at in this video.
Great Video, you must be an O'l TV repair man with all the knowledge you have, thanks for taking the time to educate some of the students of electronics out there., and please keep up the good work, I myself is can't thank you enough for the free teachings that I'm sure you had to pay for through out life. Thanks again and people out there if you don't like the video don't watch it,, it's that simple.
EXCELLENT. Once again a sober, well researched test based on science, repeatable measurement and not some air-head's "killer ears".
All it takes is some reading and LEARNING from text books. The ARRL handbook is a great primer. The math is simple, but verbal hype from the guy at the bar or the salesman at the snobby hi-fi shop will NOT teach you about this technology. Most people hate having to learn even simple science. Lazy brains I suspect.
You got that right. I sold my '66 Marshall plexi 50w years ago because if I turned it up to where it sounded good, everybody in the world hated my guts. These days, I use an old Tweed Fender Princeton and everybody's happy, including me. That's it in the little picture to your left. It works great.
I think over all, that is good advice from Dr Z for most people but I have never been one to just take things at other people's word. I need to understand and find out for myself. I am now 62 and have been this way since I was a small child. I like to know and understand from my own measurements. This attitude carries with it some risk and cost but also the possibility of much real gain. Research your subject well but don't be so timid as to never be able to Plug-It-In and Turn-It-On to learn.
I didnt understand one thing you were talking about but found this very interesting,very cool.
With correct cathode current of 40 mA on the 6L6 raised to 80-100 mA, I would say less than an hour. I used to have a McIntosh MC240 that would ruin 5881's within the day. 6L6GC would last years. If you look up the old specs of tubes you will see that a 5881 has less plate dissipation than the 6L6GC - and plate dissipation matters very much. If you can measure cathode current, just simply multiply that times plate voltage and you will get idling plate dissipation. It will increase under drive.
PS: I'd love to see a proper show-down of the 6L6 vs. EL 34 (for hi-fi purposes). Do a sweep through all extreme freqs, IM, HD, clipping, headroom, periodic harmonics, etc. E.g. do the 6L6s really have better bass? Which valve has better transients? @ which freqs? Using the same amp (output & power x-formers) with reasonable bias, cathode current & B+ adjustments makes this a 1st class objective test, Next, I'd love to see tube brands tested with no tears. -GREAT STUFF!
The Fender Bassman 300 I have a video on has around 750 volts on the plates of the 6550's. Certainly no place to plug in a 6L6 or 6CA7. 450 volts on the plates of the 6L6 is not uncommon and is typical in the old McIntosh amps. I agree with you that for a typical 6L6 style amp, 6.5Kohm output transformer performs well - much better than a 5Kohm. Thanks again.
Tubes are great, there's honesty in them, you get what you get.
Your video is really good! I learnt many things about tube replacement, and the idea of the BIAS pot on top of the chassis is interesting. I'll try all this in my amps, Thank you ;)
jibber jabber volts amps jibber. I wish i had the background to understand what you're doing. looks like really cool stuff. I used to see mcintosh amps in pawn shops 25 yrs ago. I wish i grabbed a few when they were there. Oh well.
I got here while watching vids on adjusting/modifying tube guitar amps
Fortunately my wife has stayed with me 35 years, so far. The first one didn't though :-)
As for 3 main types of distortion I assume you mean amplitude, harmonic, and frequency? And the old amps often times specified phase distortion. I will have to think about it and see if I can come up with something real. Thanks for your comments.
Pin 1 and pin 6 on the 6L6 is not used and often times used as solder connections, on the tube socket, for component connections under the chassis. The 6CA7 pin 1 is the screen grid and is most always grounded or connected to the cathode. The screen grid is internally tied to the cathode in the 6L6. So, you MUST know how the socket is wired up before swapping tubes around. A socket can be wired to take either of these tubes. Check out the pin configurations before swapping tubes. see next post
With all said below, if you can measure cathode current and limit it to about 140 mA with no drive for the pair of KT88's then it might be fun to try. EL34's draw 1.5 amps per tube (3 amps total for two) so, if your amp will support EL34's it should do OK with the filament supply for the KT88's. Reasonable cathode currents for a pair of 6L6's or EL34's is about 80 mA (40 mA per tube). Hope this helps.
I have never personally had any KT77's but they are sold as replacements for EL34. The KT66 is sold as a replacement for the 6L6. KT88's and 6550's are interchangeable. All, of course, with attention to bias and plate current when initially setup.
I can't remember the tube layout for every amp so I make mistakes. I have done mods for clients whereas I have swapped out 6L6's for KT88 and the client is happy just as you are. The potential problem is the 6L6's draw 0.9 amps each for the filaments and the KT88's draw 1.6 amps each so, most likely the filament voltage is low and the power transformer filament winding is overly stressed. It may last years or it may not. If it works and you are happy with it then all is good. Not all amps (next)
God I wish I knew valve technology like you do. Fantastic video and commentary.
I understand what you are saying but I came to realize that the creation of music and the playback of music is two different worlds and the fact that tubes are electro-mechanical devices and are affected by the vibrations of the cabinets is what allows musicians to create the music we love. Then, we try to playback at ultra-low distortion the music they created a 10% THD and above. Holding the guitar in front of the speakers may create a desirable tube related mechanical feedback that we like.
Best to tube it for what it was designed for. KT88 and EL34 draw about the same amount of filament current (1.6A and 1.5A respectfully) 6L6 is only 0.9A so if you put the bigger tubes in a 6L6 amp, you will overload the filament winding of your power transformer. 6L6 and EL34 are close in power output. You can get a little more out of the EL34 as it can take 800 volts on the plate, 500 max for the 6L6. KT88/6550 is twice the power tube of a 6L6.
In a nutshell, you have to re bias your amp if you change the tubes. Some amps have switching on the back, others the bias has to be manually done. So, the rule of thumb is just use the recommended tubes. My experience is the better the tubes the better your amp will sound. New tubes have to be broken into, they will sound very clean. I was upset when I put new tubes into my amp, it didn't have the same distortion and sounded too clean. Well, a lot to be said about old tubes and good distortion.
So, to make the socket ready to accept either the 6L6/KT88/6CA7 - you Must remove anything soldered to pin 1 of the socket and either ground pin 1 or connect it to pin 8 (the cathode) - probably best to ground it - Connecting pin 1 and pin 8 together carries a certain risk as pin 1 on metal based tubes have the metal ring connected to pin 1 so you could end up with the bias voltage on the metal ring of the tube and it will bite you if you touch it with the amp powered up. see next post
thats what is guitar players love is the distortion on a tube amp when you crank it
Yes, Tim, your question will work with a JTM 45, you can use a dozen different tubes, you do have to change the bias resistor though.
20/20 hindsight. I can only hope my retirement funds start moving soon because, according to their growth the past 2 years, i've been in some sort of time stopping bubble where nothing gains value.
I also remember nakamichi dragon tape decks in the pawn shops and i was drooling all over those too, given the choice, i could see myself getting that instead of the power amps because i was looking at all kinds of amps. There was so much to choose from compared to top shelf tape decks.
6L6's run at about 40 mA each (80 mA per pair) and it is customary to put in a fuse at twice the normal current rating. KT88's are about 70 mA per tube (140 mA per pair) so,l with that approach you would need 280 mA fuses - It would be wise to stick with the 6L6's - there are many brands out there, and 6L6's are about half price of KT88's. Hope this helps.
@MICKSHRED No, that would be a bad idea. As documented in one of my later videos, filament current of the 6L6 is 0.9A and the 6550/KT88 is 1.6A so you will overload your filament winding in the power transformer. Also, the 6L6 runs at about -40 volts whereas the KT88 runs at about -46 volts so the KT88's would be drawing much more cathode current (once again) overloading your power transformer. The other way around (might) be more acceptable.
Pin 1 on the metal based 6550's is connected to the metal base and should be grounded for safety. Pin 1 on EL34's is the third grid and must be connected to ground or the cathode (pin 8). G2 at 1Kohm sounds like a smaller tube than the 6550. G1 at 55Kohm is pretty high. Glowing hot EL34's - not good - make sure the bias voltage is adjustable to at least -60 and there is no AC ripple on it. My guess at a distance is bias supply issues. Hope this helps.
The 6V6 is a beautiful low power tube and it only requires 0.45 amps filament current (you might think of a 6V6 as 1/2 the power of the 6L6. The 6L6 requires 0.9 amps filament current. The EL34/6CA7 requires 1.5 amps filament current so it would seriously overload the filament winding of your 6V6 power transformer and I doubt the 6CA7 would even light up properly. So, unfortunately, no, this won't work.
Well, EL34's draw almost twice as much filament current as a 6L6 so the filament windings won't be overloaded and the bias is "probably" high enough (negative enough) on the EL34's so the 6L6's won't draw too much plate current. However, the EL34's can handle much higher plate voltages (800V) than the 6L6 (500V). Overall, it might work "OK" but probably best to put the EL34's back in. Google "6L6 datasheet" and "EL34 datasheet" and check it out for yourself (nice PDF from Svetlana out there).
6L6's are really good to 40 watts and 50 watts max with any reasonable amount of distortion in the creation of music. If you drive it into a square wave output you may get 60 watts. Similarly, the 6550/KT88 is outstanding at 75 watts but will do 100 watts at higher distortion. Square wave output on a KT88 might reach 120 watts. Just because someone puts 160 MPH on the speedometer of a car does not mean it will go that fast - unless it is off a cliff and it won’t be long before the ride is over.
You need to consider the plate load resistance as well, because the varying distortion and power output is a result of varying matching with the load.
If you check plate voltage with no tubes the power supply will be operating with no load and it will be abnormally high and not the voltage that will be on the plates of the tubes when they are plugged in. The bias voltage will not be affected with or without the tubes being in.
I think it's a personal preference and each tube type seems to have its own charming characteristics but overall, large and small, I am partial to the sound of the EL34/6CA7. Thanks
With an el34, you need to plan on jumpering pin#s 1 and 8 in order to get the suppressor grid polarized. Most guitar amps use open pins on the octal socket to support grid stopper and screen resistors. Since the 6l6 cathode is connected internally to the suppressor grid, no problem! Plug in a el34 into a 6l6 amp blindly might not be a terrific idea.
I think we all know about pin 6 as you mentioned, to mount a grid stopper resistor. As for pin 1, it should not be used for a tie point (I learned this the hard way) as it is connected to the metal rings of tubes like the KT88. I once used pin 1 for a screen resistor and you know what happened when I accidently touched the metal ring. So, whether it is a 6L6 or EL34 it is a good idea just to go ahead and ground pin 1. Leaving the suppressor grid open on the EL34 will probably just keep the tube from conducting very well and power will be low and distortion high. I haven't verified that but I think is what is likely to happen. Thanks for your comments.
I don`t know if you are using grid or cathode bias but check the manufacturers tube spec. Also remember that with EL34`s (6CA7) you must earth the screen pin (G3) seperately.
you look like a technician of star trek with all those weird stuff. :D
This was a great video, articulated very well! Thank you!
The extra power sure helps with bass and transients. Frequencies below 200Hz use most of your current. You speaker is 8 Ohms @ 1000Hz (NOMINAL impedance). At 80 Hz where the deep bass of the kick drum or string bass resides, most speakers go down to 1-2 Ohms -or less! -That's where all that great KT88 current goes! Now they even make a KT120. But if you get up to that power range, Why not experiement with 811s or 572Bs (they're not expensive) -You might like class A at 50 watts!
I wish I had bought every Mac amp I saw even at retail price back in the early 60's - the MC275 was $444.00 new - a NIB one now from that era would be worth an easy $10K+ - better than the stock market. I love this stuff and now that I am retired I get to play with it a lot - I don't have any vids on mods - I have done mods for musicians around my area but it seems many turn out disappointing as they don't get the sound they thought they were going to get - good luck.
Hello. Do you sell amps?
Excellent presententation. You, um, have a few more test toys than I do. You are either a serious pro or your wife has left you. Seriously, This shows that there is not much difference in the desgin for push-pull amps using the popular valves. However, it also shows that there is a more than simple grid bias to good tube life, low distortion & high efficiency. Could you do a test on power vs. 3 main types of distortion with the 4-5 main power tubes?
i had an 80 watt valve amc power amplifier using kt88 it was awesome a great sound for the money .
You should be able to plug your KT88's into your 6CA7 amp (if and only if) you have enough range in your bias adjustment to lower the cathode currents in the KT88's to 70 mA per tube. The socket as wired for the 6CA7 should be OK for the KT88 and as I mentioned below, the worst case would be bias voltage on the metal ring of the KT88's. Check your socket. If pin one is grounded there will be no bias voltage on the metal ring. Work safely with High Voltage.
Yes, I have always heard that the 6L6 is a better bass amp tube than the EL34. Not sure how to measure that. Again, got to think about it. Thanks for your comments.
It might need new tubes but It is wise to stay with the same output tubes. Your amp is designed for the EL34. ECC83/12AX7's can last a very long time as they run much cooler. Adjusting bias - that's a whole other discussion. Using a good quality contact spray cleaner does miracles and may be all you need. Dirt, grit (and beer) take a serious toll on musical amps. Spray/clean every pot/switch/and plug if you are comfortable with working on your own amp. High voltage in tube amps can be LETHAL.
Those 6CA7s tubes are very dependable and they sound great I have a Quid set in my plush amp we tinkered with they are GEs however so the tone is there. EL34s were very unstable in the plush. I did put in a presence pot in the circuit. so the amp sound like a fender on steroids a cross between English & American. I like the math of it but I am not a tech. Thank you for posting this video. c];-D
And lastly, remember that these "experiments" in swapping tubes around are not meant to be permament changes. KT88's are more powerful tubes than the 6L6 or 6CA7 and will require more energy to be delivered by the power supply. Ultimately, my goal was to answer a client's question and my last and final recommendation is to Not swap around your tubes unless you are knowledgable and experienced with working with vacuum tube amps and are willing to accept the risks to yourself and your equipment.
Thank you for your comments. There are may reasons we can't permamently swap 6550's into a 6L6 amp and one of the most fundamental reasons we can't is because of the filament current required for the 6550's. A pair of 6550's require 3.2 amps to light the filament whereas a pair of 6L6's only require 1.8 amps. The filament winding on the power transformer will be damaged over a period of time if it is overloaded by that amount.
today, all i can say is thank goodness i did not buy one because i i would need to go buy cassette tapes in the clearance shops beside VHS/Beta tape display, which is behind the film camera aisle, around the corner from the wooden wagon wheels. I still like my vinyl but it's pretty much impossible to find new stuff these days and besides that, the analog edge is gone now that mastering is pretty much all digital so tapes and vinyl makes less sense to me every day. I enjoyed the video.thanks
The 6CA7 filament draws 1.5 A and the KT88 draws 1.6 A. This probably OK as it only 200 mA difference and (probably) will not harm the power transformer. The 6CA7 and KT88 can both take 800 volts on the plate whereas the 6L6 can not. Also, the 6L6 has a filament current of only 0.9 A so putting 6CA7 and KT88 in a 6L6 amp will probably grossly exceed the filament current rating of the power transformer. see next post
I will have to think about this one. It will be a challenge. Some of my first videos are comparing KT88/6550's from vintage to modern Chinese and Russian in a Dynaco amp. You might find some of them interesting.
A pair of 6L6's is good for 50 watts and a pair of KT88's are good to 100 watts - that's just the way it is. You cannot put KT88/6550 in place of 6L6's - It will "work" but not very well and not very long. Hope this helps.
No fair! You're using a spaceship navigation computer!
(If you ever see 'Silent Running,' look for the Tektronic oscilloscope on the bridge of the ship, Valley Forge.)
Excellent video, very informative.
Thanks for showing.
KT88's draw 1.6 amps per tube filament current to a 6L6 of only 0.9 amps per tube so, that is one concern. The bias on the KT88 will need to be made considerable more negative to reduce the cathode current to an acceptable value. There is likely not enough adjustment in the bias pot to accomplish this. You will need to change bias divider resistors and even then, you may not be able to attain a proper negative voltage setting for your KT88's. You may find more help in some of my other videos.
I read an article on the KT88 tubes. Swapping from EL34 to KT88. It read that the bias feed resistors should be switched from 220k to 100k resistors to accomplish a proper reading. I believe it was something about reducing the grid resistance, so the KT88's can run at the proper rating. Is this true? and where are these bias feed resistors located?
Interesting point. Does that same ideology apply, then, to a combo - wherein the tubes are in the same box, that the speakers rattle?
Speakers being connected is only important if you are driving the amp with a signal. You can safely measure voltages with no speakers attached. High voltage is a rattle snake so connect your leads before you plug it in to the wall. Work with one hand in your pocket and never lean on the equipment and don't use both hands when measuring high voltage as you want to avoid any current flowing from arm to arm through your heart. Getting across the plate voltage is a bad experience :-(
This seems to be a very informative video -- thank you for posting it!
In the spirit of improving things, I must say that I did find it difficult to watch because of the head-mounted camera, and had to stop watching part-way through. Perhaps a tripod would help, and as would arranging your test equipment so that the DVM and scope are both near the amp can be seen with minimal camera movement. Not trying to be mean-spirited, just a suggestion for your consideration that might get you more viewers.
Hi there! Very interesting video! I am new to tube amps and just got a Peavey 5150. Eurotubes says that if you install an integrated quad set of 2 JJ KT88's or 2 JJ KT66's and 2 JJ 6L6's then the amp does not need to be biased and it's basically "plug and play". They said they did a lot of research on this and that it not only sounds great but saves you from having to bias the amp. They recommend you put the 2 KT's on the outside sockets and the 2 JJ 6L6's in the 2 middle sockets. You seem to know what you're talking about so I was just wondering if you think this is safe and is a good idea. I know this is an old video so I hope to get a response cause I just ordered the tubes today! Thanks!
I have only experienced the KT120 in some Dynaco spin-offs - can't remember the brand - impressive tube, very healthy bottle. I think their cathode current is a bit higher than the KT88 and KT90 so. some care should be taken before switching them.
Even though It's done everyday, I was always afraid to put EL34's in a 6L6/6550/KTxx/ socket due to a slightly different pinout. 6L6 and the like use a common (7AC) socket where pin 1 is not connected (NC) and pin 8 is connected to the cathode (k) and grid 3 (g3) where as the EL34/6CA7 (8ET) socket uses pin 1 connected to grid 3 (g3) and pin 8 is connected to the cathode (k) on its own.
So what youre saying is that we can put kt88's in place of el34's, we just have to make bias adjustments to the tubes specs.
Hello. Just catching this video for the 1st time. Great video. You keep referring as 6L6 but arent those 6L6GC? Theres a big difference between to the 2.
For a RPN calculator it is 1 (enter) 1 (+) to get the display of 2
Is that where the stainless McIntosh (repos) are coming from? I haven't seen them on Ebay yet but I have seen the "redone/rechassis" amps that look great. If someone starts making repo Mc transformers I have serious concerns that they will be the quality of a Mc transformer which is the heart and soul of the Mc performance. Thank for bring this up.
Brilliant video man!
Now there is a handy and safe tube puller / inserter available on ebay, item 254242318237 Easy, simple and safe. A perfect tool for any audiophile or vintage electronics technician.
Interesting. Makes perfect sense, really.
Great Vid!! ✌😎
Those ajustments you make by changing only the cathode resistance??
First amp shown in this video is a Dynaco Mk III, not a Mk IV. The Mk IV was a 40-watt chassis which ran two EL34s.
Thanks for noticing and correcting.
I never noticed and now that TH-cam has taken away the "bubble" type comment that I used to be able to make and correct mistakes I can no longer make changes to my videos.
I am not familiar with your amp so I can't say for sure. 6L6 & KT66 are interchangeable as is the EL34 & KT77 but there is considerable difference between 6L6 & El34 and again, a considerable difference with the KT88/6550 - basically, a KT88 is capable of twice the output power of a 6L6. EL34's have a much higher plate voltage rating than the 6L6. Without a considerable bias voltage range the three tubes above can not be substituted one for the other. There is also filament current issues.
I don't have my tube specks book but the diference on these tubes are the plate resistance I know that you could get up to 70 w from a pair of 6550's and only 50 w from a pair of 6L6's of coure adjusting the bias for AB1 mode but if the output transformer is not matched to the plate resistance that I believe is 6.5 Kohms for the 6L6 you are not going to have optimum efficiency and I don't remember which amp but it had 6L6 with 475 volt plate voltage most run the 6L6 at 350 to 400 plate volts.
I too am contemplating a KT88/6550 tube change in a older marshall 1988 for more clean headroom, none of the new amps have the tone i’m after and I can’t afford a Dumble or Bludotone!
+R. Weaver It could be OK and you may like the change in tone. If the Marshall runs EL34's the filament current change to the 6550 will be OK. The filament of the 6550/KT88 and EL34 draw about 1.5 amps each whereas the 6L6 only requires 0.9 amp each) What I hope was noticed in the video is that the bias voltage for a 6L6 is approx -37 volts, for a EL34 approx -44 volts and for a 6550/KT88 approx -55 volts ( all approximately). What the point is, if your Marshall will allow you to adjust the bias a bit more negative, the 6550 could be OK but you must check it or you will just ruin tubes. Another point is, with a 6550 amp with a bias of approx -55 volts, even with the higher plate voltage in the EL34/6550 amp, you can, for a short time, put a 6L6 in the socket and it won't go up in flames since the 6L6 typically needs only -37 volts bias to set the cathode current at a safe level. But, in a 6L6 amp with -37 volts negative bias if a 6550/KT88 is placed in the sockets, the tube will draw Way too much plate/cathode current and things will not go well. Also, placing either EL34 or 6550 in a 6L6 amp will overload the filament circuit of the power transformer. Good luck with your project.
ElPaso TubeAmps Tranny is shorted at present, so changing to accommodate tube change should help. I am currently looking to change mid control to enhance 500k a little more.
+ElPaso TubeAmps unless I'm mistaken I should be able to modify the circuit so as to get the bias where needed
soo?? this second amp has a B+ of around 350v when the 88s load it down...in order to qualify 70ma each. i am looking to convert a marshall superlead (34s) to 88s.. could you verify the needed change in splitter resistors and screen resistors...?
I would like to try EL34s in my Scott 296. Can it be done, replacing GE7581A(KT66) with 6CA7/EL34s? The 7581 is similar to 6L6GC with more output 35w vs 30w respectively.
May I ask why you switched amplifiers in the midst of what would have been a great comparisom. Unless the B+. screen, OT, are exactly the same there is no ability to compare. Additionally the 6L6s were not equal as indicated by the difference in getter oxidation.
question, when your figuring out the wattage output of an amp, i know you use voltage squared divided by the ohms gives you the wattage, but when your checking the voltage do you have to have to use a true rms meter, or is there a way to do it with a regular multi meter?also, do you know of any cheap devices that I can get to verify the actual output of an amp before clipping? or do you basically have to have a osciloscope, frequency driver, rms volt meter, resistive load(to emulate speaker), and distortion analyzer? or can i not use an osciliscope and just get away using a distortion analyzer?(somehow i have a feeling a dist analyzer is more expensive then a scope)
sorry for my igrorance but i'm relatively new at electronics repair and am trying to set myself up for working on tube/ss power amps.
You have described pretty much the whole setup. A distortion analyzer and an audio spectrum analyzer is probably the most expensive things to have. An oscilloscope, a signal generator, a RMS reading volt meter, and dummy loads are necessary. You can drive the amp up to just below clipping and get a pretty good idea of its performance with a THD around 1%. Noticing the shape of the sine wave is important also. Pretty much any volt-ohm meter will work for reading RMS voltage at 1 KHz like the vintage Triplett and Simpson analog meters Knowing the frequency response of the meter before purchasing is a good idea and ideally you would want one that is flat from at least 20 to 20 KHz or better. The Tektronix TM500 series instruments will let you have an oscillator, voltmeter, distortion analyzer all in one mainframe. The most expensive items for this type of setup would be a SG505 oscillator and the AA501 distortion analyzer - you could pay $1K for those items but possibly less. You can buy good used Tektronix scopes for little money and in my opinion they are definitely the best to have. Of course new ones with a spectrum analyzer and oscilloscope built together (called Mixed Domain Oscilloscopes - MDO) are lots of thousands of $$$. Good luck with your enterprise. If there is anyway I can help, please let me know.
So you could potentially run a pair of 6L6GC's with a pair of KT88's? Or am I mistaken?
How easy would it be to take a 6l6 push pull pentode configuration to a 6l6 triode configuration.
You did not mention the 6550 with I liked better than kt66 or KR88 and my favorite was 8417
have a bias supply with enough range to be able to make the bias supply negative enough to move from 6L6's to KT88's. 6L6's need bias something in the -40 volt range whereas the KT88's need bias voltage in the -50 volt range. I don't know about your Mesa Boogie Triple Rectifier - some Mesa Boogie's don't have adjustable bias voltage - you will just have to investigate the schematic carefully to determine if it will work and not overheat.
hello sir this video is really teaching a lot and it does help me a lot I still reading and comprehending more about tube amplifiers, I have a question if you don't mind, I still in the process of making a KT88 push pull ultra linear hi fi amp "not for guitar" home use on the cathode of the KT88 what resistor should I use 10R 3W is fine or should I get a 5W ? many people have use 5W and 3W what is correct ?
the B+ will be around 460V DC from a Edcor transformer model XPWR247-120 330 - 0 - 330 you mention already "long time ago" that I should use 4.2K output transformer one more question about output transformer, should I get one that can go up to 100W or one that supply about 60W ? I was thinking about the B+ and I'm not sure what transformer should I get because of the 460V DC B+ I'm a bit confuse to be honest
Regards
Juan
Hi Juan - here is the way I would look at it - absolute maximum cathode current for a pair of KT88's is about 275 mA - that is the current if the tube is being driven to maximum sustained sine wave power. (from a 1962 RCA tube manual) That current flowing through the 10 ohm cathode resistor would dissipate power of (I^2*R) which would be about 0.76 watts so a three watt resistor would be OK - even a 1 watt resistor would not burn up but would get pretty hot. Another possible way to look at it is, in case something goes terribly wrong in the tube and there is an excessive amount of current flowing through the tube, a 3 watt resistor would burn up before a 5 watt resistor would and "might" help protect the output transformer from excessive over current. So, we can see that making everything larger is not necessarily a good way to go. On the other hand, if the primary of the power transformer is fused properly, then it should blow and protect the amplifier. Another way to protect the circuitry is with a fuse in the high voltage line - a 0.5 A should be sufficient. Many guitar amplifiers have a HT fuse. Once you complete the amp, if you have one of the handy "Kill-A-Watt" meters that you can get pretty cheap at Lowes' you can actually measure the power up current of the amp and set the line fuse to some value above that point, somewhere around twice the max current that you measure. Hope this helps.
ElPaso TubeAmps
good idea about the HT fuse and the cathode resistor to be 3W and yes this does helps about the "Kill-A-Watt" really good idea I didn't think about that, I'm going slow with the build so when I have something done I will let you know thanks for the help :D and yes I have been aware of high B+ dangers don't worried I be really careful my MOS was 94F Special Electronics Device Repair you have a great day sir :D
Regards
Juan
how much voltage do these amps run on the plates? I'm guessing around 420-450?
Have you try to use KT120 as well?? It seems like this tube relatively quite new to market...
Can i Replace in a Marshall 1986 or 1987 the EL34 with the KT88-94 for More clean Headroom ?
The Heater current of the KT88 is about the same of the EL34 ....I know i have to lower the bias feed resistor from 220k to 100k to avoid tube problems ..the question is : the transformer can handle the bias current draw with the KT88... i have read that the tube should be biased like a EL34...but if bias this way they should be biased too cold !!!!! Can i bias at 60% with a plate voltage of 470v-480v without overheating the transformer ?
+Giuseppe Loriga You are right about the filament current being close in the EL34 and 6550/KT88 so that should be OK. Plate voltage on both tubes can be as high as 800 volts. I doubt it will damage your transformers but I would limit the plate current in each KT88 to 70 mA (140 mA for both KT88's) by adjusting the grid bias. As the amplifier is built for an EL34 the bias is probably not negative enough for the KT88 so it will have to be adjustable to a more negative voltage. Pin 1 of the EL34 is g3 and is probably grounded at the socket. Pin 1 of the KT88 is attached to the metal ring of the 6550/KT88 - make Sure pin 1 of the KT88 socket is grounded to the chassis - as it probably already is..
+ElPaso TubeAmps Thanks for your Help ... So what is the effect of the reduced plate current in the tube ? Can i bias the tube at 60% ( 43.8ma with a plate volge of 480v Class AB Fixed Bias) or it will run colder ? Last question the pin1 in my el34 tube socket is grounded to the chassis through the 1 Ohm resistor , is that ok for the KT88 ?
This is interesting. Nice Vid!
Always use the tubes recommended for the correct voltage, bias, and primary impedance of the output transformer. Period.
It's all about the pre amp section, also I don't get why Guitar players insist having the Amp head on top of the speaker stack. Man all the 12AX7's will go microphonic so fast compared to a Audio system that isn't going through a Earth Quake of vibration.. Kills me seeing nice OLD Tubes going into Guitar Amps sitting on the stacks!