Need some help from the tube gurus

[hurdy_gurdyman]

My trusty Scott LK-48-B has a problem. The right output tranny has been running a lot hotter than the left (never hot enough to burn flesh, but very uncomfortable after running all day.) The left tranny is only warm. I have checked volages. The tube plates all have 410 VDC, the cathode resistors have 0.3 VDC on them. All four resistors are exactly 10 ohms. No DC offset between tubes. Both sides appear to be identical.

Last night I did some serious comparing between the sound of the two channels. The right channel (hotter one) has much more prominant high frequencies than the left. I checked my EV LS-8 speakers with an old Sansui 8080 receiver I have and they sound identical to each other, so not a speaker problem. I'm using a quad set of JJ EL84's. I tried switching them with no difference.

I guess I need advice where to go from here. I have a couple pairs of trannies for 7591 tubes, but nothing for 7189/EL84 type tubes (if it turns out to be the trannie.) Don't know if they would make a decent substitute or not, but I suspect not.

What else can I check before giving up?

Dave :/

[NOSValves]

Dave,

The right transformer on all Scott amps will run hotter the the left simply because its surrounded by heat sources that the left tranny doesn't experience. It just has allot less ventilation to dissipate heat. The 5AR4 is fairly close which is a huge heat producer the iron just has a natural absorption effect. The transformer being hotter has most likely nothing to do with the high frequency out of the right channel. You say it's more prominent out of the right does this mean you think its improper or out of wack in the right or could it be the the left is duller or has a lower basic volume output then it should be? Being a Kit amp this could be a cold solder joint almost anywhere it would take a scope to narrow down whether its in the front end, tone controls, switches, phase splitter or output section.

What condition is this amp in? Partially rebuilt, completely rebuilt or bone stock? What are the values of the cathode resistors? Is there just one per channel like original?

One thing is for sure 410V on the plate is correct for the amp but those JJ Telsa EL84's were never meant for it. They will work but they won't last long! That is a true 7189 based amp you have is not an EL84 amplifier. Some vintage EL84's can take that heat but the modern ones don't care for it at all. But still this wouldn't be the probelm most likely.

If you want give me a call and will do some mild trouble shooting over the phone 810-287-6135

Craig

[hurdy_gurdyman]

Thanks for the info, Craig. I replaced all the coupling caps a while back with orange drops. I'll recheck the solder joints. In my opinion, the left "duller" sounding channel is the one that sounds wrong. The highs are just plain dull. I'll play with it for a couple of days, and if I can't fix it, maybe I'll call. I have a scope here.

Dave

[NOSValves]

Dave,

What about those cathode resistor values? You should actually measure there value in Ohms with the amp off. They often drift over the years. I take it you still have the originals in there? The schematic shows them as 10 ohm shared between the 2 output tubes. If they are indeed 10 Ohm than you actually have the amp biased to low is should be biased to .44 to .50 volts per CHANNEL. If you happened to have installed a 10 ohm per TUBE then the proper bias is .23 to .25 per tube. I'm not sure where you got this .30 volts figure from?

Wooo sorry.... I just read your first post and you do indeed have 4 cathode resistors. The proper bias for amp is .25 max! Your pushing all 3 of those transformers pretty hard at .30 volts. If you happen to have 4 spare output tube you may want to try them. If those EL84 have been idleing at that setting for long I bet there shot. Running EL84's at 410V on the plates is way about the 350V spec then couple that with running the plate dissapation above the absolute max is not a good thing. I still will be surprised if this solves your channel imbalance but running the amp like this is asking for a transformer failure. I bet the power transfomer is getting pretty toasty warm also. Where did you ever get the .30 figure for biasing this amplifier?

Craig

[hurdy_gurdyman]

Craig,

I worded my first post a bit wrong. There are four resistors, but they are in pairs, thus actually being one resistor per channel. The 10 ohm resistors have been bypassed with a second resistor (don't recall what they are at the moment) but they measure right at 10 ohms (or as close as my old Archer meter can tell.) As for the .3 volts, someone elsewhere told me to keep the voltage below .5 volts, and if the tranny felt to hot, to go below .4 volts. Again, my poor old meter is using it's 2.5 volt (analog) scale and may be off a bit. I was worried about the right channel tranny burning up, so I used a voltage on the lower side to be safe (maybe I'm paranoid, but I've lost 3 output trannies on various amps through the years, including a nice Stancor ultra-linear.) The tranny seems to run about the same temp anyway, so I'll increase it to about .45 and see what happens.

I went and re-soldered all the new caps (I replaced all the coupling caps in the amp and preamp.) I admittedly don't have much high frequency hearing anymore, but it sounds like the two channels are closer to sounding the same, but I need to listen at a time when I'm not so sleepy.

How can one tell if an output transformer is too hot? After running this for 12 hours or so I can place my hand on it, but it's awfully uncomfotable. Doesn't actually burn, though.

Dave

[NOSValves]

Dave,

So what your saying is each channel is running two 10 Ohm resistors in parallel this would equal having a single 5 ohm resistor. What these means is all the figures I have given you for having a single 10 Ohm resistor should be cut in half. So you are running the amp to hot. The max I would set the bias for each channel is .25V and in all reality with EL84 tubes I wouldn't run it over .22. I'm not sure where you got the information from but I would be very careful in the future.

The formula is pretty simple the 222D and LK48B should be biased at 44ma per pair of output tubes per Scott spec. so you take this figure

.044 x value of resistor = bias voltage setting

.044 x 5 = .22VDC for 5 ohm resistor value

.044 x 10 = .44VDC for a 10 ohm and so on.

*These above figures are for a pair of tubes*

The maximum idle plate dissipation of a EL84 is 12 watts to figure that you take the current the cathode is run at time the plate voltage

410V X .O44 = 18.04 /2 = 9.02 watts per tube.

You do not want to run the Max plate voltage way over spec and then also push the tube to its ragged edge on plate dissipation. The max suggested plate voltage of a EL84 is in the mid 300 range.

Craig

[NOSValves]

DP

[hurdy_gurdyman]

Dave,

So what your saying is each channel is running two 10 Ohm resistors in parallel this would equal having a single 5 ohm resistor. What these means is all the figures I have given you for having a single 10 Ohm resistor should be cut in half. So you are running the amp to hot. The max I would set the bias for each channel is .25V and in all reality with EL84 tubes I wouldn't run it over .22. I'm not sure where you got the information from but I would be very careful in the future.

The formula is pretty simple the 222D and LK48B should be biased at 44ma per pair of output tubes per Scott spec. so you take this figure

.044 x value of resistor = bias voltage setting

.044 x 5 = .22VDC for 5 ohm resistor value

.044 x 10 = .44VDC for a 10 ohm and so on.

*These above figures are for a pair of tubes*

The maximum idle plate dissipation of a EL84 is 12 watts to figure that you take the current the cathode is run at time the plate voltage

410V X .O44 = 18.04 /2 = 9.02 watts per tube.

You do not want to run the Max plate voltage way over spec and then also push the tube to its ragged edge on plate dissipation. The max suggested plate voltage of a EL84 is in the mid 300 range.

Craig

Craig,

The two resistors on each side are not both 10 ohms. The one 10 ohm on each side measures about 12 ohms, so another resistor of higher value has been added (memory is telling me they are around 40 or 60 ohms) to bring the total resistor value of each side to 10 ohms. In effect, there is exactly 10 ohms measured for each pair of output tubes, which is as it should be.

Thje JJ EL84 tubes have performed flawlessly for over 10 months in this amp without even a hiccup.

I did a resistance check on both output trannies and each side measures the same, so at least I'm happy knowing that the trannies are ok. I'm sure the problem (less highs on the left) is a solvable problem, probably in the preamp section. Kind of sounds like a light blanket covering the left speaker.

I'm going to temporarily dig out my old Fisher 500C receiver and use it for a while until I figure out what I'm going to do with the Scott. What I really want is an old Heathkit AA-121. I like those old Heathkits, so maybe I'll watch for one and sell the Scott.

Anyone want a Scott LK-48-B? Needs a little work, but is fixable. No power supply hum to speak of and has new orange drops coupling caps. Would consider trade for Heathkit AA-151 or SA-2.

BTW, the Fisher is sounding pretty good right now, just a bit ugly to look at.

Dave

[NOSValves]

Dave,

So your value is exactly 10 Ohms per channel. Then your biasing the amp on the cool side good enough. Yes I'm sure with the bias set that low the EL84 will leave a reasonable life. But the amp would sound better with 7189's and the bias cranked up a good measure. I for the life of me can't figure out why someone would tag on a bunch of resistors to correct a drifted value. Much easier to just install a new 10 Ohm IMHO.

Good Luck

Craig

[hurdy_gurdyman]

>I for the life of me can't figure out why someone would tag on a bunch of resistors to correct a drifted value.<

One had already been corrected before I aquired the amp, so I simply played with a couple different resistors untill I got the other channel to have a 10 ohm reading. I didn't have any exact replacements on hand, anyway.

I may have found the main problem. I finally just sit down and started comparing resistance values between right and left channels until I found a substantial difference. There is a 68 ohm resistor connected to the coupling capacitor in the stage just before the final coupling cap. One of them (right channel the one running hottest) is open. Of course, I don't have that particular value on hand, so I'm going to check some local tv repair shops tomorrow and see what I can find.

Dave

[Tom Mobley]

very informative thread. Thanks for both of you for running this down right here on the forum where everybody interested can read and learn.

[hurdy_gurdyman]

Sucess! Replacing the 68 ohm resitor fixed the problem. Turns out the resistor was in the negitive feedback circuit. Not a good place for an open resistor. Both channels sound identical now and I have the bias set right.

Dave

[NOSValves]

Yup that would do it! Good work.

[hurdy_gurdyman]

Carig,

After several hours of running, the sound is still excellent, but that right tranny still gets way hotter than the left, uncomfortably hot. Maybe it's normal and I'm to much of a worry wort. While I was working on it earlier, I checked carefully the resistance of the right channel tranny vs the left one. They are identical, so no short. Wasn't able to find any more bad resistors. I put my hand on it for about 15 seconds then had to pull off as it was starting to hurt. I can leave my hand on the left on as long as I like.

Does this sound like normal extra heating from the rectifier?

Dave

[NOSValves]

Well its depends on just how wimpy your hands are [] You know I work on this stuff day in and day out before doing this for a living I spent 20 years working home improvement so my hand are pretty well toughened up. If you work at a desk and never do manual labor your hands could be pretty sensitive compared to mine. As you can see the hand touch method is not very scientific or accurate. I will say that the right transformer always runs a good bit hotter then the left. If you cathode resistors are indeed accurate and you have the bias set properly then there is not much more you can do. Well if you really wanted to get crazy you could swap the transformers but a seriously think it wouldn't change a thing.
Just thought of one thing how are you balancing the output tubes in each channel?

Craig

[hurdy_gurdyman]

Craig,

My hands are pretty soft these days, as I've been working as a luthier and musician for several years. Because of health issues, I don't even do a lot of either these days.

The only way I know how to balance for DC is with a VTVM or VOM by checking for any DC between the output pair grids. Chances are there is a better way, huh?

Dave

[hurdy_gurdyman]

Craig,

I found some instrutions on setting balance by hooking a wire to the right channel speaker output and plugging into left channel tuner input, remove right driver tube and adjust pot for minimum hum. Decided to try this with both channels and see if the heat lessens. I'll leave it on a few hours and see.

I'll be leaving this afternoon for a music fest and won't be back till Sunday evening. I'll post how the amp is doing before I leave, then check back in Monday morning.

Thanks for the help.

Dave

[hurdy_gurdyman]

Looks like I solved the heat problem. I cheated! I took one of those little pentium cooling fans (nice little ball bearing model) and hooked it up to a 6.5 VDC wal-wart (normally runs on 12 VDV.) The fan, which now runs very slow and quiet, is mounted next to the test sockets on top of the chassis, pointed diagonally toward the transformer (the side facing the 5AR4 rectifier tube.) There is barely any noticable airflow, but it's just enough to cause the two trannies to run at about the same temperature. It's beginning to look like it may have been heat from the rectifier and power transformer was causing the hot running of the right tranny.

The amp is sounding great and is not running hot enough to worry me anymore.

Thanks again Craig for your patience and help.

Dave