Changing/Biasing Tubes for the first time in JCM 900 SL-X

[samth3mancgp]

Hey everybody. I bought a JCM 900 SL-X about a month ago. I love the way it sounds, but I am pretty sure that it still has the stock tubes in it, and would sound much better with some new ones. Right now it has all Marhsall ECC83 preamp tubes and some Sovtek 5881s in it. I thought about switching to EL34s because that is a pretty popular change on them, but I also read of many people that prefer them with 5881s, or 6L6GC so I thought I would make my life easy by just getting a matched quad of JJ 6L6GC power tubes and a mix match of different types of preamp tubes to do my first full tube change.

I know that the preamp tubes are basically a free-for-all and I can move those around with all different kinds for the perfect tone, but the power tubes are a little bit more difficult in that they need to be biased every time they are put in the amp. I have already done a decent amount of reading on other forums about changing tubes, and spent a good amount of time on the Eurotubes website watching their videos.

Here are some pictures of what I have to work with.

New tubes:


A view from above:


A view from below. (I read that you can stick a 1ohm/1watt/1% tolerance resistor between pins 1 and 8 that I can then use to measure with my multimeter. Right now there is just a wire bridging that gap as seen in the pic. I don't NEED one of those bias probes right? I would really rather not have to buy one. I have several digital multi-meters.


Lastly, closeup of the preamp tubes:


and closeup of the power tubes:


It would be very appreciated if somebody could help me through this. I have made at least 50 cables, built my own mic preamps, and tore apart tape machines and other gear many times. I know I can do this, and I'm not afraid to do it. I just need a bit of guidance for my first time doing it

[Jazz P Bass]

Really, if the amp sounds good, why not leave it alone?
But if you must, sigh....
Using the 1 ohm resistor is a good ploy.
By measuring the voltage across it, you can extrapolate the current going through the tube at idle.
So if you measure , say, 35 millivolts, that indicates that 35 milliamps is coarsing through the tube at idle.
By measuring each tube you can come up with a balance between the two halves by swapping the tubes around.
Please be careful in there.
High voltage can ruin your day.

[g1]

1 ohm resistors at each power tube is the best way. However, you can also remove the "OP valve fail" fuse and clip an ammeter across it. This will give you the current for a pair of tubes, so you won't know how well they are matched. I suppose you could test 2 at a time for matching purposes.

[samth3mancgp]

1 ohm resistors at each power tube is the best way. However, you can also remove the "OP valve fail" fuse and clip an ammeter across it. This will give you the current for a pair of tubes, so you won't know how well they are matched. I suppose you could test 2 at a time for matching purposes.

Could you explain this in more detail? I see two of these fuses on the back of my amp. If I could get away without putting those dang resistors on the tube sockets then that would be great.. The only ones I could find here in town were 1ohm, 10 watt, 5% tolerance resistors. And even though I am pretty good with a soldering iron, this sounds easier overall.

-Sam

[g1]

To reiterate,

High voltage can ruin your day.

It can also end your life.
To bias via the OP valve fuses you need a multimeter set to measure DC milliamps (approx. 200mA range). You remove the fuse and clip your meter to the 2 wires connected to the fuse holder. The reading will be for a pair of output tubes, so divide the number by 2 and you will know what the current is through a single tube (assuming they are reasonably matched). Now do the same for the other OP valve fuse.
Be very careful with the meter in current function. Often the probes have to be plugged into different jacks. In current mode the meter probes are connected like a straight piece of wire. If you forget to change the jacks back and try to measure voltage or something you will blow stuff up.
Now you have measured the idle bias current, you multiply that by the plate voltage (pin 3 DC volts) and you know your idle wattage.

[samth3mancgp]

Take One... Old Tubes

Okay I just ran through all of this using my OLD tubes just to get my hands dirty. I basically followed the steps from this thread from another forum, except that instead of measuring a single power tube across the 1ohm resistor, I was measuring from the power tube fuse area (combined milliamps from two power tubes). I measured 480V from pin 3 with the tubes in, and 76ma from the fuse area. Half of that would be 38ma/tube. So 480v(.038 amps) was 18.24 for my idle wattage?

The last step in the thread I linked to said that you can then mess with the bias trim and run the tubes a little bit hotter or cooler as long as it was "operating within capacity (usually 20 watts, but varies depending on the tube)". I ended up adjusting up to 78ma (39ma/tube) but I don't know what idle wattage is recommended or safe for the tubes. The JJ 6L6GC matched quad that I bought says 19.8 on a little sticker on the side of each one. Is that the maximum idle wattage that they can take? Around where should I adjust the trim for these new tubes to idle at? I don't have a clue how well matched the old 5881s are. Probably not very well since when I first measured the fuse slots one measured 75ma while the other measured 63ma.

Is the reason for measuring BOTH fuse areas simply to make sure that the tubes match? Then I can move them around between their slots to get the closest match in milliamps between the four? As long as they are close to matching, I only use 1 of the measurements to calculate where the bias needs to be?

Hopefully all of these numbers that I am throwing around are not shocking (pun intended ) to you guys and I am doing something right. I haven't jolted myself yet and I don't plan to! I've done it before with other things and I know it sucks! Sorry for the mess of questions guys. I really want to be able to understand and do this myself. I'm just waiting for the day that a band comes in to record and his amp blows up and I am able to fix it right there in front of them. They would immediately become a return client!

[g1]

Basically you have it all correct. Your idle wattage is 18.24. Russian 5881's are considered 30W max. tubes so you are biased at about 60% (18/30). You will hear recommendations ranging from anywhere between 50 and 80 % (some recommend 50% max. when plate voltage is over approx.450V). Remember that your plate voltage (pin3) will change as the bias is changed.

The JJ 6L6GC matched quad that I bought says 19.8 on a little sticker on the side of each one. Is that the maximum idle wattage that they can take?

Forget about the sticker. It has to do with their matching process and is only meaningful to their test procedure. The only use to you would be if you needed to buy more tubes from them and wanted some that match what you have now. You could ask for the same number.

Is the reason for measuring BOTH fuse areas simply to make sure that the tubes match? Then I can move them around between their slots to get the closest match in milliamps between the four? As long as they are close to matching, I only use 1 of the measurements to calculate where the bias needs to be?

Yes to all. Also, only the fuse where you are measuring should be removed. The other should be installed. If you were to do it with both fuses removed, 2 tubes would not be conducting and your plate voltage would rise, throwing your readings off a bit.
Now that you have grasped the concept, consider that many amps don't have fuses for the power tubes. The same concept is sometimes used with the mains fuse, sometimes called "line draw" biasing. Here you measure the AC current at the mains fuse. For example, once you have your amp biased the way you want, you would check the AC amps through the mains fuse holder (& line voltage). Again, for example, say you measured 1.5A and the line voltage is 122V. Your line draw is 1.5x122 or 183 watts. Next time you bias for 183W line draw and you are done.

[samth3mancgp]

Basically you have it all correct. Your idle wattage is 18.24. Russian 5881's are considered 30W max. tubes so you are biased at about 60% (18/30). You will hear recommendations ranging from anywhere between 50 and 80 % (some recommend 50% max. when plate voltage is over approx.450V). Remember that your plate voltage (pin3) will change as the bias is changed.


Forget about the sticker. It has to do with their matching process and is only meaningful to their test procedure. The only use to you would be if you needed to buy more tubes from them and wanted some that match what you have now. You could ask for the same number.


Yes to all. Also, only the fuse where you are measuring should be removed. The other should be installed. If you were to do it with both fuses removed, 2 tubes would not be conducting and your plate voltage would rise, throwing your readings off a bit.
Now that you have grasped the concept, consider that many amps don't have fuses for the power tubes. The same concept is sometimes used with the mains fuse, sometimes called "line draw" biasing. Here you measure the AC current at the mains fuse. For example, once you have your amp biased the way you want, you would check the AC amps through the mains fuse holder (& line voltage). Again, for example, say you measured 1.5A and the line voltage is 122V. Your line draw is 1.5x122 or 183 watts. Next time you bias for 183W line draw and you are done.

Awesome, so I got it right! So the question now is are the JJ 6L6GC tubes that I bought also 30 watts max or are they 25? Now I have to go in and run this whole procedure again with the new tubes inside. It seems like I am basically just doing this biasing thing to make sure that the tubes aren't running too hot or cold, but after that I can just adjust it to taste somewhere between 50-80%.

When I was measuring through the fuses I was only removing one at a time

Regarding what you said about the line draw biasing.. I think if I were to bias an amp that didn't have the easy way out measuring the power tube fuses, I'd probably just do the 1ohm resistor technique. Although that line draw technique would be really nice if the bias trim was on the outside of the amp! Thanks for your help, this ended up not being too difficult!

[g1]

The JJ is 30W (listed under limiting values) here: JJ-6L6GC.pdf
Here's the beauty of the line draw method, there is a device called "kill-a-watt" (around $25), which you plug into the outlet, then plug your electrical device into. It will tell you the voltage, current, and wattage of the device connected, so you don't have to remove the chassis to clip in your ammeter. Now consider many old Fender amps have the bias pot accessible without removing the chassis. So you do your bias the long way the first time, then write down the reading on the kill-a-watt. Next time you can bias without removing the chassis.
There is another (more dangerous) bias method called "shunt method" which you can read about here: Bias FAQ

[samth3mancgp]

That's pretty cool! I have read about the shunt method, that's the one that uses the transformer right? A lot of people seem to be scared of that one because there is more chance of getting shocked?

Alright, so I have ALL NEW tubes in my amp now! I measured 475v as the plate voltage after installing the new JJ 6L6GC tubes. I measured 79.4ma across power tubes 2 and 3 (fuse 1) and 76.5ma across 1 and 4 (fuse 2). Are those numbers supposed to vary a bit? I didn't swap the tubes around to find a better match.. I used the average idle current of the two fuses (77.5ma) for my calculation. Half of that being roughly 38.75ma/tube. so 475v(.0387)/(30)=61% of 30watts. I ended up not touching the bias trim at all to get that calculation. That's technically within spec right? Does biasing the amp sometimes result in no need to adjust anything? I know I could sit around and adjust to taste, but I like the way the amp sounds. It seems like it has a slightly rounder and full bottom end/low mids.

After all this I did a preamp tube shootout.. I recorded some dry guitar into ProTools to re-amp with different configurations of preamp tubes. I miced up the cab with an AKG C414. I first sent out the clip and recorded the tone using the stock preamp tubes (Marshall ECC83 in V1-V3, and then a Ruby in V4). After that I switched all of the preamp tubes out with new ones. (TungSol in V1, JJ in V2, and EH in V3 and V4). I re-amped the same clip through that configuration. I compared the clips and the only difference was the new preamp tubes/configuration has a tiny bit more brightness to it and clarity. VERY subtle difference. After that I inverted the polarity of one of them so I could hear only the artifacts that were different between the two. It was all staticy distortion artifiacts in the high-mids (> 3khz). A tiny bit of difference in the low end. I meant for it to be as scientific of a test as possible. I knew that I could spend all day swapping tubes around and playing the amp. The differences are so subtle that you need to do a test like this to perceive them properly, and not fall for placebo. (The new tubes are NEW and HAVE to sound better, so my brain makes me hear a difference). I just left the new preamp tube combination in there and didn't swap anything else around. No amount of swapping will make or break the tone. Anybody interested in hearing the clips?

Overall this was a great experience and I have a much better understanding of all of this. Thank you all for your help!

[g1]

Sounds like they are pretty evenly matched and you are correct, sometimes no bias adjustment is required. If you are at 60% and happy with the tone then that is great. A bit more or less (and the preamp tube tone) is a matter of personal taste. What is the correct amount of mustard for a hot dog?

[Enzo]

70%


Polish sausage is good at up to 80%.

[Jazz P Bass]

Brown or Yellow mustard.