Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: pbman1953 on January 19, 2018, 08:42:50 am
-
I installed 6- KT-120's in my Fender Pro300. The amp's stock tubes call for 6-6550's. After talking to some , I was told that the amp could handle the KT's. After playing it for about 5 minutes, with a moderate setting, one of the protection fuses blew. I wanted to see if the set the blew the fuse were ok, so I refused that set and removed a fuse from a previously used set. The set that made the fuse blew were fine. The amp continued on the balance of the night with 2 sets (4-tubes).
My question is if anyone has tried the KT120's in their 300T or Pro300?
I sent along the bias page from the manual to show how I used the builtin controls to bias.
-
Fuses are designed for over-current protection. If you had the proper fuse installed and it burned and installed another with 4 tubes and it ran fine, I would guess you had an over-current issue.
If you know the amperage the Power Transformer, HT, and Heaters, did you check to see if you exceeded the heaters with the 120's. These take more heater current than 6550 and you added it times 6. Also, did you get a large voltage drop on your plates of the 120's?
You can check for current easily in the amp. Please post the specifications of your Power Transformer.
-
Thanks Ed, I'll see if Fender will divulge the specs. They can be kind of tight lipped at times
Thanks
-
They can, but you can email Patrick at Mercury Magnetics for the spec for an exact replacement and it will be the same.
Go here:
http://www.mercurymagnetics.com/pages/mainframe.htm#_misc/contact.htm
Click the center email as it is Patrick. Stock replacement specs only, not one of their overbuilt jobs. It usualy takes him about 5 minutes to reply, but they are in Cali and we are east coast, so keep that in mind.
-
Sent,
Thanks
-
Here's the spec sheet from Patrick
-
The attachment doesn't have amperage which is needed. Send him another email and tell him. Just say you have this PT and you are considering changing from 6550 to kt120. You want to insure you have enough filament current.
-
I'll have it Monday
-
It looks like the tube protection fuses are 250ma for the combined cathodes of each pair of output tubes, if this is the right schematic: http://el34world.com/charts/Schematics/files/fender/Fender_Bassman_300.pdf
While the high heater current is a potential worry, it shouldn't be the cause of those fuses blowing.
If these tubes are properly biased, the output transformer impedance should limit the full-load power draw of KT120s tubes to about the same as the 6550s would draw, so this shouldn't be happening.
If the pairs are significantly mismatched, one set could be biasing too hot (and another too cold) even though the combination of all 6 tubes overall biases correctly.
Also, you need to let the amp warm up "fully" before biasing. I've been using some KT90s and was surprised by how long it takes them to heat up just enough to put out sound, let alone for stable/consistent bias, compared to EL34, 6L6, 6V6, etc. There's debate about how long tubes should be allowed to heat before bias "settles", but in my opinion I'd monitor bias until it's been on for about 15 minutes. If you turned it on, biased it as soon as it was hot enough to read the current, then played for 5 minutes... the problem was likely that bias current was continuing to increase throughout that time.
-
Can I do your steps and bias with the back panel controls? I didn't plan to take the amp out of the case yet. It's heavy
Thanks
-
Yes, I looked for a proper schematic and couldn't find one with fuses shown. It's frustrating when posters don't post the schematic. It's unclear: i) if the fuses are original to the amp; and, ii) if they're affecting filament, or cathode current.
My question is if anyone has tried the KT120's in their 300T or Pro300?
Personal experience is nice, but unnecessary. This is a general math problem that can be addressed by people of this Forum, but only with the basic info.
-
I'll find it and scan back
-
Here's a link to the service manual-
http://www.stevehouse.com/bucket/Fender_SUNN_300t-11x17.pdf (http://www.stevehouse.com/bucket/Fender_SUNN_300t-11x17.pdf)
Pages 16-21 have it all.
They are listed as screen fuses.
It was too big to upload here
-
It appears that the fuses in question are F5, F6 & F7 @ 100mA ea.; and ea. serving the screens of 2 tubes. It does not seem possible that 2 normal screens, whether 6550 or KT-120, could possibly draw anywhere near 100mA. Looking at the tube specs, both tube type's G2 max out at about 13mA x 2= 26mA. This suggests that something other than good KT-120's may be causing blown fuses.
You may wish to measure screen current @ idle & under max signal conditions. Run the amp into a suitable dummyload, and measure the voltage drop across the screen resistors. Use Ohm's Law to calculate the current flow. This should verify screen current.
Always suspect tubes first. Because this is happening after warm-up, a possible tube fault is that internal tube elements are heating-up, expanding, and shorting out. I think only a short could cause enough current flow to blow these fuses.
-
I didn't get the PT specs yet but I was chatting with my tube supplier and he chimed in. He made some sense as far as chance taking but it's well beyond any warranty, age wise. Plus both transformers never seems to get hot , warm at best.
100mA for two big power tubes isn't that much actually. Putting in the KT120's might push past that. If you can find fuses that are a little bigger that might be fine. You are putting in larger power tubes, you have to expect there could be more current through the circuit. You could try 125mA fuses and see how it goes. I'm sure that will void any warranty, so be aware of that. And of course you are experimenting at your own expense. I can't take any responsibility for what happens. You should monitor the temps of the transformers to be sure they are safe.
-
Specs came in-
282 B+ 1.2 amps
250 ½ an amp
Filament 12 amps
40V 1 amp
-
100mA for two big power tubes isn't that much actually.
This is not for the power tubes; just for the screens.
Putting in the KT120's might push past that.
No need to guess. I already did the math from the tube specs. I already showed how to measure screen current. Note that with an LED in the screen circuit, thee might be as much as say, another 30mA through the 100mA fuse. That would kick current through the fuse up to maybe 56mA; whereby a 100mA fuse makes sense.
If you can find fuses that are a little bigger that might be fine. You are putting in larger power tubes, you have to expect there could be more current through the circuit.
Once again, no need to guess.
You could try 125mA fuses and see how it goes. I'm sure that will void any warranty, so be aware of that. And of course you are experimenting at your own expense. I can't take any responsibility for what happens. You should monitor the temps of the transformers to be sure they are safe.
The screen + plate supplies are drawn from the same secondary winding. That behemoth PT is likely to shrug-off another few hundred mA of current draw (which I don't believe is there anyway). It's the tube screens that are delicate, but maybe a chance worth taking. I still say there's a short or partial short somewhere. Either inside a tube, or maybe somewhere in the complicated screen supply circuit. Subject to actual measurement, it does not appear that the KT-120 screens are drawing more current than the 6550 screens; nor near enough current to blow a 100mA fuse.
BTW: are these slo-blo fuses?
-
JJ, you may find this interesting. I have a push pull UL running a pair and just finished putting it on a dummyload to see how the screens react to cranking the amp up. Idle each is around 8.? Ma, just a little over 8.
Max clean is actually more than I previously remembered. Diode rectified and 565 plate HT current 320 ma. UL taps 40%. 4.2k Max clean rocks the screen per tube 38ma. Consider the onset of distortion and may even see more.
I have read others measure 50ma at 100v which is rediculoous . If you must give up some wattage to lower the screen the kt120 will be happy.
Sure the datasheet gives the idea you may get 150 watts from a pair, but if treated Ampeg SVT 6550 these run well.
-
As already been said, the math doesn't add up. Blaming the fuse is inviting trouble. Also as already been said, always look at the tubes first. You said you ran four just fine for the rest of the night. Did you swap positions with the "bad" set or leave them in the same holes? Unless the fuse was a no-name Chinese fuse I can tell you that the quality control of any U.S. Fuse company is flawless. It's not the fuse' fault. You had an over current event. Upping the fuse size is an invitation to let some magic smoke out.
Ed, what are you running for screen resistors?
Jim
-
I was waiting for 2 tubes and was running the amp with 2 sets. Lets call the them slots 1 & 2. I added the 3rd set , in slot 3, and the fuse blew. I pulled the fuse on slot 2 and installed a new few to slot 3,with the new tubes. The amp ran fine . Hence running with slot 1 & 3 and all was fine.
-
Now that IS odd. Something with the bias circuit?
It originally had 6550"s? A full sextet? Never had any issues with them?
Jim
-
I added the 3rd set , in slot 3, and the fuse blew.
Which specific fuse blew?
Max clean rocks the screen per tube 38ma.
Yes, measurement is key, (maybe I figured wrong?). Ed's 38mA x 2 = 76mA screen current + LED current might blow a 100mA fuse.
Still I thin k it's OK to up the value of these fuses. The tranny should be fine. Tube might blow, but what the heck it's for science!
BTW, the SS circuitry around the 3 fuses culminates in a 16V output. Where does that go? Does it trip a relay to shut down the amp in case of screen failure, or what?
-
The only blown fuse was the protection fuse that's in front of each tube pair
.
There some internal fuses and the main fuse.
-
As already been said, the math doesn't add up. Blaming the fuse is inviting trouble. Also as already been said, always look at the tubes first. You said you ran four just fine for the rest of the night. Did you swap positions with the "bad" set or leave them in the same holes? Unless the fuse was a no-name Chinese fuse I can tell you that the quality control of any U.S. Fuse company is flawless. It's not the fuse' fault. You had an over current event. Upping the fuse size is an invitation to let some magic smoke out.
Ed, what are you running for screen resistors?
Jim
1.5k right at the moment. Nothing like Hifi.
Screen current is variable or am I crazy? I mean I am watching the current through a meter. When I first started messing with these tubes everyone was really hyped as we had a new tube stronger with extended bass. And it is true to a point. They are not KT tubes. I have had them laying beside them real GEC kt88 and tungsol 6550 the old ones and the ge hotrodded 6550 and they look similar to the ge but just bigger.
I am not sure I have read anyone who doesn't just substitute with them to either hear them or in hopes they will run in old amps like SVT Ampegs.
-
The only blown fuse was the protection fuse that's in front of each tube pair
Did all three fuses blow? If not, which one(s)? (this ties-in to Ritchie200's suggestion to move tubes around)
Screen current is variable or am I crazy?
??? You already posted different screen currents under different signals.
-
The only fuse , and there's only been one is the protection fuse, outside and directly in front of a tube pair.
I mentioned before that I replaced the bad fuse and removed a fuse from the center pair. Hence running 4 tubes. The tubes that had the blown fuse worked fine with a new fuse the balance of the night.
I didn't have to move tubes, I made active tubes inactive.
-
FYI-
I may test the bias again.
When I first measured it, the bias wouldn’t get to the .1 (100mv) the manual asked for. The most I could get was .098mv and the pot was maxed.
Ed suggested to have the bias supply resistor tweaked to get more range.
-
The only blown fuse was the protection fuse that's in front of each tube pair
Did all three fuses blow? If not, which one(s)? (this ties-in to Ritchie200's suggestion to move tubes around)
Screen current is variable or am I crazy?
??? You already posted different screen currents under different signals.
Yes, I know. Sorry rhetorical comment. Should have stated such. What I was gettig at Ritchie200, who plays with a pig, was inquiring about screen resistors. I got the idea he was thinking peak 38ma was excessive, which it may be. It is what it is. I am finding more and more builders who are using these are regulating screen current.
For science, I agree. Up those babys to 125ma.
Also, the poster emailed me with some info you guys should know. With the 120's installed, he loses range of bias which means it is loading a bit more. He said he could only get them biased by turning the trimmers all the way. Since idealy the negative voltage should be near center.
I looked at the schematic for a couple of minutes, then lost my mind. I would guess it is either a overcurrent or a bad tube. Great conclusion, I know. :dontknow:
-
FYI-
I may test the bias again.
When I first measured it, the bias wouldn’t get to the .1 (100mv) the manual asked for. The most I could get was .098mv and the pot was maxed.
Ed suggested to have the bias supply resistor tweaked to get more range.
I suggested if you plan to run these tubes you need to get a proper bias travel.
It is up to you if you want to to try running the tubes. I said I would try it if I really wanted to use them.
-
When you look at the output tube schematic, you will see each pair of tube screens connects to LED1, LED2, or LED3. In this case LED1 does not refer to a LIGHT EMITTING DIODE. It's just a label, although a confusing label. LED1 simply connects to point LED1 in the SS monitor circuit and of course, the fuse. So, if you been wasting time looking for light emitting diodes (LED1, LED2, LED3), stop. There ain't one!
BTW, the SS circuitry around the 3 fuses culminates in a 16V output. Where does that go? Does it trip a relay to shut down the amp in case of screen failure, or what?
That 16v is the B+ input to the SS circuit. It ain't an output. All the SS circuit does is lite a green LED if you have screen voltage, ie, the fuse ain't blown. If the fuse blows, the green LED turns off and the red LED turns on to indicate the fuse is blown. Green light good. Red light bad.
Under normal operation screen voltage is applied through a big voltage divider to the base of the transistor. The transistor is just a switch and will be turned on, which also turns on the green LED. When the transistor is on, current is shunted away from the red LED and it will be off. Now, if the screen fuse blows or you lose screen B+ for any other reason (including putting the amp in standby mode), the transistor will switch off, causing the green LED to turn off. Since the transistor is off it will no longer be shunting current away from the red LED and it will turn on.
This is a quote from the user manual...
On the top side of the chassis, at the rear of each pair (total = 3) of output tubes, is an LED and a fuse. The LED glows green
when the associated pair of tubes is operating properly, and glows red if a tube failure has caused the related fuse to blow. If a
fuse blows, replace it only with one of the same type and rating. NEVER use a fuse with a higher current rating, as this could
damage the equipment and present a serious safety hazard. Next, replace the tubes in their respective locations, check and
adjust the bias and balance.
NOTE: it is normal for the LEDs to glow red when the amp is in standby mode.
-
The only fuse , and there's only been one is the protection fuse, outside and directly in front of a tube pair.
I mentioned before that I replaced the bad fuse and removed a fuse from the center pair. Hence running 4 tubes. The tubes that had the blown fuse worked fine with a new fuse the balance of the night.
I didn't have to move tubes, I made active tubes inactive.
Yes, you did. A bit hard to keep track. This suggests maybe a fault in the section of SS circuitry where the fuse is blowing.
@ sluckey: thanks for clearing up those side issues.
@ed:
What I was gettig at Ritchie200, who plays with a pig,
So that's the problem; he needs to get a puppy! :icon_biggrin:
Bias: overall tube bias should be checked in a normal way -- not just per the Manual instructions by way of the the screen connection.
For science, I agree. Up those babys to 125ma.
For added protection, a fuse could be temporarily added to ea power tube, say at the cathode. This would definitely protect the OT and might save a tube from self-immolation.
-
:think1: Here's an idea... Put a 6-pack of 6550s in it and leave it alone.
-
If I upped the bias supply resistor I'd need some help to determine if its R333, 334 or 335
-
If I upped the bias supply resistor I'd need some help to determine if its R333, 334 or 335
Like I said, I lost my mind looking at the schematic. Opamps, transistors and microprocessors and a SS loop. I can see a Tube Screamer in there tho. :laugh:
I believe your Bias circuit is more complex than just changing one resistor. I am guessing because I do not want to try to figure it out unless I had the amp in hand. I would pull the chassis. Install the 6550, make readings. Install the 120 and take readings to see how much voltage drop you get. Check the 16volt too. Like I mentioned, I am not sure how the bias works, but I did look long enough to see collectors on transistors.
Anyway, to INCREASE negative voltage you want to reduce the resistor, if there is only one to reduce.
-
:think1: Here's an idea... Put a 6-pack of 6550s in it and leave it alone.
This is the easiest solution you will find. Now back to work.
-
You must have all six tubes installed and working with good fuses before you set the bias.
Your bias circuit is really just two identical bias circuits, one for each side of the push/pull triplets. Whatever you do to R333 and R334, you must also do to R353 and R369.
-
> In this case LED1 does not refer to a LIGHT EMITTING DIODE. It's just a label, although a confusing label.
Yes, confusing.
It is the feed from the fuse, one per pair. But voltage (implying fuse status good/blown) is indirectly monitored by a "LIGHT EMITTING DIODE" as in the quoted text. Here's an analysis.
The scheme is perhaps over-complex and has "flaws". They have to tell you Red is normal in Standby. (Logical if you know how it works and recognize they did not spend extra pennies to shut-down this branch of the 16V when the hi-voltage is switched off.) It would also show "bad" for wiring faults. But the LED color should tell you, usually, when a fuse has blown.
I have *never* seen *only* a screen fuse(s) with monitoring. This suggests the amp is volted and loaded way close to the edge of the cliff where the screens just vaporize.
-
Correct, the LED is red on stand by and green during operation.
-
the LED is red
the 1st time I installed and tested a Phillips MRI the table didn't work, table control board had a red LED lit, spent 4hrs figuring out Phillips used RED LEDS to indicate NORMAL power on :cussing: :think1:
-
What color is an LED that ISN"T THERE? :BangHead:
@ Shooter: newfangled electronics are even worse. Sometimes a red light means the appliance is ON; sometimes it means its OFF but plugged in. Sometimes a blue light means its ON; or maybe an amber light means its ON. Gives new meaning to: The lights are on, but nobody's home!
In an old Woody Allen movie he's convinced that his appliances get together and meet in the living room to conspire against him when he's not home. I know how he feels.
-
Read all about it in the user/service manual. :icon_biggrin:
-
Read all about it
that's where the 4hrs was spent, have you ever read Dinglish :BangHead:
-
So Ed, is it overcurrent or a bad tube..... :think1:
:icon_biggrin:
Ed brought up a good point that others are also alluding to. Put the 6550's back in? Fortunately Ed is uniquely qualified to answer this question. Ed, you have spent a lot of time with this tube. Do you think it is worth pursuing? In this application they would be loafing along and would probably last forever. Big bold clean tube with extended lows would seem to be ideal. What do you think? Worth the effort?
Jim
-
So Ed, is it overcurrent or a bad tube..... :think1:
:icon_biggrin:
Ed brought up a good point that others are also alluding to. Put the 6550's back in? Fortunately Ed is uniquely qualified to answer this question. Ed, you have spent a lot of time with this tube. Do you think it is worth pursuing? In this application they would be loafing along and would probably last forever. Big bold clean tube with extended lows would seem to be ideal. What do you think? Worth the effort?
Jim
I really like the tubes. I have not built anything with the that is not Ultra Linear. I have also never used more than 2 on Push Pull. I have built 3 amps very similar, just one blockes more bass. Smalller caps. All SS rectified and none with NFB. Tried it and the amp gets stiff, very stiff. Mine was first and the play time is not that great, but the other 2 are used by guys who run them harder. One is a Steel Player and the other has a large house gig on Guitar. The Guitar just get a retube and his were over 3 years old.
Most simply say provided your filiment current can handle the tha addtional current, you are good. Well a Leslie 147 Iron has plenty, but if you install 120's in it and get the correct cathode resistor to get close to 90% dissipation the voltage drop is too much.
What I will say is I would never install any of these in an amp that has collector value and run the risk of trashing an original Transformer. The problem is most that install these things never check operational conditions. The simply add up the heater current and it is the same (or even if it is a little over some guy will ALWAYS say transformers are generally overbuilt) will put them in.
Here is where I will try and have had good results. I have re-tubed an AMPEG SVT, 1972 with these. Checked bias and current. These things were a huge difference since the tubes he had were very old.
Use them and make proper mesurements, certaily. Stick them in anythiing that has is supposed to have enough current, no way.
Amps are designed to a tube and not the other way around. Sometimes we can substitute and it is fine, but even a simple substitution will cause a component change if it is just a bias resistor or adjust a bias pot. Same thing. The more the tube changes the circuit conditions, the more that is needed to be changed.
Just set on on a table beside a Real 6550 and ask yourself if you think adding 6 of thses things to my amp is going to chage the way the circuit operates. In your Marshall, since I do not know the value of the amp this would guide me. I would still try them provided I had monitored the amp with the KT88's to compare to the 120's. If conditions changed where I had to do more than change bias, I pull them. If not, you will end up chasing more problems.
-
Also, I think PRR may know more about these tubes than he has mentioned. I have read a lot of his posts everywhere on 6550's.
Also, Dummyload had a plan to bake one or 2. Maybe these guys can share their understanding of these. I know the Data Sheet is crap. Also, I only know 1 tube vendor who test over 500 volts. Most use a Maximatcher at 400 volts max. A tube can change a lot between 400 vdc and 550 vdc.
Yse the tubes would be loafing on plate voltage, but do we know if the transformers would be. This is why I run 2 of them with 320ma, 400-0-400.
-
I know nothing about this new KT120. It may be a 1.5X slice of KT88 guts; or it may be less than it seems.
The amp is already a beast. I don't think anybody but the player (maybe) can hear what tube is in it. Unless some newfangled oversize bottle gives it troubles (and screen fuses should NOT be blowing). _I_ would fill it with Sovtek 6550 and go to work. The limited abuse I have done says they are/were as well-made as GE-Kentucky and did not creak (on dummy-load obviously!) like vintage Tung-Sol.
-
I've given some thought to this based on the post. What it really is that I'm really upset that SED is gone and their 6550's are triple the price for anyone who ha them.
That's what is really bugging me. :BangHead:
So I'm thinking to go back to a set of Tung Sol reissue 6550 or possibly a set of JJ - KT-88. Or maybe a st of Gold Lion, I've never used them yet.
Another factor s that New Sensor has all these tubes and are they really different or are they same with different labels? :w2: They bug me.
At this time, I've been using the KT120's in my Sunn 2000s and it's fine. Plus, I remember, using my Bias Rite on the 300T, months back. The OT transformer (with JJ 6550's) was putting out over 700 volts and with Sunn the KT's around 514- 520. Could this be the reason the amp has trouble with the KT-120's? With them the voltage has to be much higher right?
I guess I could get up the muscle to pull the amp out of the cab and check it.
-
Over 700V? On the plates? With JJ's? I know the 2000's were supposed to be conservatively rated at 120W but the old Super Leads were pushing that with 150V less. Maybe the new KT88's are built crappier than the 6550's, but there are very few brands of KT88's that will last in my amp at 620V. I realize how you beat the screens is also a factor as there is more to it than just plate voltage - but wow.
Jim
-
Jim
Just to be sure -
Over 700 V with the JJ's(6550) in the 300T
Low 500's with the KT120 in the Sunn 2000s
-
Just my opinion, but it looks like the KT120s are drawing excessive current or possibly V4 my be causing problems with the bias voltage. B+ should be 742 bolts at idle per the schematic. Have you measured the DC bias voltage for the control grids? It should be about -50 volts (-49.4 per schematic). I would measure the bias voltage at each tube socket. In comparing the datasheets between the 6550 and the KT120, the current draw at idle shouldn't between the two tubes shouldn't be enough to cause that much of a difference in plate voltage. Operating screen grid currents between the tubes don't differ significantly per the datasheets.
Where did you get the tubes? Some vendors do a better job of screening to weed out culls than others.
-
Also, looking at the schematic the combined plate current and screen grid current for 3 tubes is 100 milliamps per side.
-
TC Tubes, St. Paul
-
I have no experience with that vendor. Looking at their posted tube testing methodology, there is no mention of tube burn-in. Tubes can drift significantly in the first 100 hours or so. Most premature failures due to manufacturing defects occur during this time period.
-
http://tctubes.com/
-
I took the amp to the bench and pulled it out to test with my Bias rite-
Here are the first ,before tweaking results, based on the Fender manual-Settings are supposed to be bias at .1 and balance at 0mv
Voltage and cathode readings from tubes 1-6-
749/32 757/ 68 757/ 30 773/ 64.6 768/ 39.7 780/ 67.8
Then after using the balance and bias control to try and smooth things out
781/ 44 770/ 41.3 774/40 780/ 33 773/ 40.8 781/ 33.1
Retested factory points from the the back panel controls-
Bias- 113.7mv
Balance- .4mv
For the first part of this the OT was cold and the bell top of the PT was slightly warm. At this point the OT is still cold and the PT is warm all over, with the bell top warmer than the sides. I'd estimate with a playing session the PT would be hot but not burn you.
Overall I see the balance playing a big part to try and close the gap. But tubes 4 & 6 are just going along for the ride.
-
Here's a little functional diagram that shows how to quickly adjust bias and balance and also which tubes are affected by each pot. While desirable, it's not so important for each tube to conduct the same amount of current. The important thing is for the top triplet of tubes to conduct the same ***total*** current as the bottom triplet.
I also included the screen fuses so you can quickly tell which tubes will be affected if a particular fuse blows.
-
Very cool diagram, thanks!
I just finished at least 45 minutes here at home playing along with some tunes at a pretty good clip. I can probably 60-70% of a gig level volume and all the fuses haven't popped and the amps sounds killer.
So I need some opinion on the factory settings vs. what I came up with at the end and what I should do with the 2 tubes that are at around 40% of dissipation.
OT is slightly warm, slightly
PT sides are twice as warm of the OT, the bell top is slightly less warm than the sides.
.'
-
While desirable, it's not so important for each tube to conduct the same amount of current. The important thing is for the top triplet of tubes to conduct the same ***total*** current as the bottom triplet.
Set the bias according to the factory setting. 100mV per side. Your settings are kinda hot.
-
If II do that I'll have 3 tubes at almost full spec dissipation and fuses will
-
Retested factory points from the the back panel controls-
Bias- 113.7mv
Balance- .4mv
According to what you said earlier, you're running over spec. Back it down to 100mV per side!
-
The retested was after the tweaking to bring down the current.
If look back at my post, the first part was the factory settings. The second set was after i played with the bias and balance to get the cathode close together.
-
You're not listening. You just said earlier that the bias is set for 113mV. THAT'S TOO HOT!.
Connect your meter probes to J9 and J10. Adjust the bias pot until the meter reads 100mV. That's the factory setup.
Now connect the meter probes to J10 and J11. Adjust the balance pot until the meter reads 0mV. That's the factory setup. (And just for shits and grins, connect your meter to J9 and J11. I bet it will read 100mV just like J9 and J10 reads.)
Your amp will now be running cooler than it is while set to 113mV.
-
Just to be sure that we're on the same page. Do you know what the Bias Rite meter is from Weber VST, Ted Weber's company
-
YES I DO. NOW PUT IT UNDER THE BENCH, GRAB YOUR METER, AND SET THE BIAS AND BALANCE THE WAY THE FACTORY SAYS.
Do you understand that functional drawing I made for you?
-
I see the bias side and the balance side.
Last week I installed the fifth and sixth tube then factory balanced. The bias would not go to 100 but to . 98. The balance would go to .1. After that the fuse blew.
If i put it back LIKE YOU WANTED, I fear the fuse might blow again.
-
Just set it the way you want it.
-
Bias Rite meter
If I'm not mistaking, the meter works with1 tube at a time, but the pot controls 3 tubes, and the TP's monitor 3 tubes.
If you have 113mA and you can't adjust to 100mA, you have a problem, since 100 < 113.
-
The 100 millivolt reading is a combined measurement for each bank of 3 tubes. These readings should be taken using a Volt/Ohm meter as Sluckey recommended. Also, you should measure the bias voltage at TP42 (Pins 3 and 8 of tube V4) as indicated in the schematic. This should be at least -49.4 volts. It may need to be more negative, -50 volts or more to get the KT120 tubes to idle in the correct bias range.
-
:hijack1:
Could I "cheat" by installing an Rk to "individually" tweak to taste
back to regular programing
thx
-
Looking at the schematic, the Balance Control (R361) adjusts the bias voltage for tubes 6, 8, and 10. The Bias Control (R335) adjusts the bias voltage for tubes 5, 7, and 9. I can see where you might blow a few fuses if you were trying to get 100 millivolts between test points J10 and J11.
As a starting point, adjust the Bias Control to get the -49.4 volts on pin 3 of V4. Then adjust the Balance Control to get the -49.4 volts at pin 8 of V4.
Put each tube independently into the V5 socket and measure the current draw between J9 and J10. Do not make any Bias or Balance adjustments while doing this. Write down the current measurements for each tube. It should be in the neighborhood of 33 milliamps or so for each tube. On paper, sort the tubes by individual current draw and assign them into two triplets of tubes, the goal being to best balance the aggregate current draw between the two triplets. Let tubes V5, V7, and V9 be triplet A, and tubes V6, V8, and V10 be triplet B.
Put each triplet of the tubes into their respective tube sockets and measure the current draw between test points J10 and J9, and test points J9 and J11. These two measurements should be approximately equal.
-
I know you are nervous about adjusting the bias IAW the factory instructions. I would be too if I had just bought 8 KT-120s to go into an amp designed for 6550s. So I wont ask you to turn either bias or balance pot.
I would like for you to simply check two voltages though just so we can get an idea how the amp is operating. Let the amp warm up a few minutes. Connect your black meter probe to J9. (This is chassis ground.) Connect the red probe to J10. We expect about +100mV. Now move the red probe to J11 (don't move the black probe). We also expect about +100mV.
What two voltage measurements did you get?
-
I did what you said and I'm getting to what I reported before,
Bias- 113.7mvBalance- .4mv
Then, I was able to bring it back to manual spec of bias @ 100 mv and balance at 0.
What's next?
-
You're done. Tubes should last longer now that you are running them cooler. I still want you to do this and report back...
I would like for you to simply check two voltages though just so we can get an idea how the amp is operating. Let the amp warm up a few minutes. Connect your black meter probe to J9. (This is chassis ground.) Connect the red probe to J10. We expect about +100mV. Now move the red probe to J11 (don't move the black probe). We also expect about +100mV.
What two voltage measurements did you get?
-
Someone asked me to check my fuse type and I looked up what I had which was a Buss GMA100MA- It's a fast acting.
Seems I have to order the T version which is a slow blow.
Off to Amazon
-
That was me.
-
This falls back on me because I didn't have the correct fuse. :BangHead:
New fuses have been ordered.
Sorry for taking everyone down this rabbit hole. I learned a lot though.
Thanks!
-
There was a simple point I wanted to make but you just would never answer me. I wanted you to know that having 100mV at J10 and 0mV between J10 and J11 is exactly the same as having 100mV at J10 and 100mV at J11.
The amp really has two identical and completely independent bias circuits. Calling one a balance circuit is kinda misleading. You adjust the bias pot to set the top three tubes to 100mV as monitored across a 1Ω cathode resistor at J10. Then you adjust the balance pot to set the bottom three tubes to 100mV as monitored across a 1Ω cathode resistor at J11. Now since you have the same voltage at J10 and J11, there must be zero volts between J10 and J11.
-
Thanks but this was more of a matter of understanding not a non willing to cooperate.
I appreciate every time all of you spend the time to help.
-
It's usually easier to understand when you cooperate.
Really doesn't matter at this point. We all learned something.
-
Last question in this Bias/ Balance type setup-
At what percent of dissipation would the setting be?
Just curious, to know if this has a numbered answer
Thanks
-
Percent of dissipation would be actual dissipation divided by maximum dissipation multiplied by 100.
-
Good to go! :thumbsup:
-
Percent of dissipation would be actual dissipation divided by maximum dissipation multiplied by 100.
Am I supposed to take an average of all 6 tubes then use your method above?
-
Look at a data sheet for KT120. Note the maximum plate dissipation. It may be 60 watts. So, 60 watts x 6 tubes = 360 watts max.
Then measure the plate voltage for the top triplet of tubes. Maybe it's 750 volts. So, 750V x .1A = 75 watts for the top triplet of tubes.
Then measure the plate voltage for the bottom triplet of tubes. Maybe it's 750 volts. So, 750V x .1A = 75 watts for the bottom triplet of tubes.
And, 75 watts + 75 watts = 150 watts for total power of all six tubes.
Percent of dissipation would be actual dissipation divided by max dissipation times 100%. So, 150W/360W = 0.416 x 100% = 41.6%.
If you need to know the % dissipation for each individual tube use your bias rite to get the plate voltage and plate current for each tube and do the math.
-
Seems like I'm looking pretty good in the middle of the 60-70% range.
All 6 dissipations add up to 233
233 -/- 360 = . 647
.647x 100 = 64.7
-
I'd love to see the numbers you used to get 233.
I used these numbers that you provided when you had the bias set at 113mV and I come up with a total of 180.25 watts.
Then after using the balance and bias control to try and smooth things out
781/ 44 770/ 41.3 774/40 780/ 33 773/ 40.8 781/ 33.1
Retested factory points from the the back panel controls-
Bias- 113.7mv
Balance- .4mv
So, now that you have properly set the bias test point to 100mV your numbers should be less than 180.25 watts.
Would you post the numbers you used to calculate 233 watts?
-
I think there is some confusion here, or at least I am, ha.
1)
This set below was taken from the start of all this. Taken after I performed the factory instructed settings- Bias at 100mv and balance at .0
That's how I came up with 233
Formula used and I'll round up each number : voltage x cathode- Ex- 749 x .032 = 24
749/32 757/ 68 757/ 30 773/ 64.6 768/ 39.7 780/ 67.8
24 . 52 23 50 . 31 . 53 . - 233
2)
The first test above (factory book setting) . had every other tube off from each other. So I decided , with using both pots to get them closer to each other.
This set was taken after I played/ tweaked the bias and balance pots to get 179. I used the combination of the 2 pots to level out the readings.
781/ 44 770/ 41.3 774/40 780/ 33 773/ 40.8 781/ 33.1
35 32 31 26 32 26 182(adjusted number)
-
I used the combination of the 2 pots to level out the readings.
What voltage is on J10 and J11?
-
At this point J10 is 100 and J11 is 0.
Just a point to mention- When I was tweaking the the bias /balance for the second set of measurements, I moved some of the tubes to different positions. Maybe that could of helped. Because in the very beginning I couldn't get 100 at the bias . The pot was maxed out and only went to .098 on my lowest meter scale. Now I seem to have more swing and the pot is not maxing.
-
What voltage do you measure between J11 and chassis ground?
-
Chassis ground as in J9?
If yes, 0
-
Chassis ground as in J9?
If yes, 0
Yes, J9 is chassis ground. And no, J11 cannot be zero. Humor me. Just stick your red meter lead into J11 and touch your black lead to chassis ground. It better not be zero. What does your meter read?
-
I warmed up the amp for a least 15-20 min.
The bias reads- J10- 100.4
The balance reads- J11- .4
DC scale set at 200m, my lowest setting
-
Chassis ground as in J9?
Chassis ground literally means the metal body of the chassis. So you can put, or clip, your ground probe to the chassis.
Some circuit point might be grounded to the chassis, so you could use that. But you don't know how perfect that other ground point is, unless you do a continuity check between it > chassis body.
-
Ok, I get that, so I clipped to one of the spring holders that hold down the tubes.
Very different results-
Bias- J10- 5.2
Balance- J11- 103.5
Like it's backwards compared to using the book method
Another thing I tested- The spring mounts totally ground out to each other and zero out on my meter. The J9 to chassis ground has a measuruament of 3.6 ohms. Not a total ground out. Hmm, what's with that?
-
Bias- J10- 5.2
Can't be so.
You just don't get what I'm driving at. I know it. That's OK. We wont talk about it.
-
Keep trying to take measurements Sluckey is telling you. I watched you use a meter and begin to rely on or at least try. Sure we are a bunch of old hound dogs that don't get to run anymore. Makes us grumpy. :icon_biggrin:
It is clicking, but sometimes we get a block and cannot understand. Just ask specifically and Sluckey being the fine gentleman, tends to be direct. Wanna know a secret? He loves teaching this.
I am glad you got them running, that is a lot of dollars.
-
OK, I admit it , I may have seemed like a block head but I think , with some help, and you know who you are, that I've figured it out
Here's the setting after setting to the manual-
J10 to J9- 100mv
J11 to J9- 99.6mv
------------------------------
J10 to chassis ground- 107.6
J11 to chassis ground- 105
-
Yes, there's some resistance from test point J9 to to actual chassis ground. This would explain for the slightly different voltage readings. Presumably the manufacturer knows this and has accounted for that in the instructions to use J9 as a test point @ 100mV. The difference in mV between J9 vs. the actual chassis is small; only a few mV which = mA due to the 1 Ohm bias sense resistors. The bias is reasonably within balance. Again this is the bias specified for 6X 6550's -- 3 tubes per side; ea bias sense R serving 3 tubes. The 100mV bias specification is not necessarily correct for KT-120's; though supposedly KT-120's should be plug & play (except maybe for heater current demands as someone said earlier).
-
Last night , with the new fuses, the amp performed like a champ. It kicked butt!
A couple questions come to mind-
If 100 is set for 6550's, what would be the test range for KT120's?
Is higher than 100 a hotter or cooler bias?
I remembered to get a feel for the PT last night before I shut it down. The bell top was hot and it stung and I couldn't leave my hand on it. The sides were less in temp than the top.
Thanks to all! I do appreciate the help.
-
Is higher than 100 a hotter or cooler bias?
Higher is HOTTER, 100 is current.
what's good for 6550's is also good for the 120's in your amp since it was designed for 6550. If your PT is taking skin, you have maxed out the amps capability.
-
Ouch, def, no skin just stinging hot.
Thee's so much power maybe 'll try a lower number to cool things off a bit.
I'm sure there's no harm . right.
-
there's no harm . right.
Your amp was designed to dissipate 100mA per side, the amp doesn't care whether that "LOAD" is from 6550s, 120's, or a BIG A :cussing: S resistor. the "extra" heat is probably from filaments.
play it til I breaks
-
Treat it like a Harley Sportster. Fix it till it's broke.:laugh:
Seriously, the amp circuit was designed to idle at 100 milliamps with 6550s. IMO, you have the amp set up as close you're going to get to optimum using KT120s.
-
pbman, I know you have used this tube before and apparently liked the result. Now that you have had a chance to wring them out in this amp, what do you think? How does it differ from the 6550's? Give us your sonic assessment!
Jim
-
1. How hot is the tranny with 6550's compared to KT-120's? IOW you're not giving a baseline for comparison.
2. If the tranny is hotter with KT-120's don't automatically assume it's HT draw. A KT-120 heater is rated to draw up to 1.95A; 6550: 1.6A. 6X KT-120 heaters could be drawing about 2A more in heater current. This point has been made more than once. That alone could account for more heat. It would be helpful to measure these things rather than speculate.
Tubes should not be re-biased in accordance with tranny temperature! This is absurd.
Hot trannies are no problem unless so hot as to cause failure. Your tubes are too hot to touch too. In the unlikely event that the tranny really is is too hot to run KT-120 tubes, then either: i) don't use those tubes; ii) install a fan to cool the tranny; iii) remove the bell covers for better heat dissipation; or iv) install a supplemental filament tranny.
-
I don't have any 6550 vs. KT120 , tranny heat tests. I didn't pay as close attention back then. I do now because the KT was not the stock tube. One thing is the ambiant heat around the the 120's compared o the 65's is more.
A fan would probably be a very good idea. I'll look for one to face it because, I won't mount one inside.
-
If I cannot keep my hand on a transformer without it burning me, I would be concerned. JJ mentioned you not having a comparison. Next time you play, put the 6550s back in. It should not take you but a few minutes now that you know how.
This amp is really not the same as an old SUNN or even the Traynor you have. I wouldn't simply expect it to easily run these tubes unless at the time of the design it was considered.
If you do this, you will know for sure. I play these tubes, I get it. They are thunderous even with guitar.
I would just hate to see a transformer burned when it is so easy to swap tubes and check. There are other ways to test, but this one is easy.
Doesn't mean you cannot run them. You know how to install a filament transformer. A fan is not a solution to stressing a transformer, but is a good thing.
-
I will Ed, either over the weekend or the next time we play here, which is in 2 weeks
Thanks
-
You can use an infrared thermometer. I got one for about $17 at Harbor Freight.
This is a rare occurrence where I disagree with Ed. 1. I don't believe a Tranny that's hot enough to burn you is necessarily a problem. 450 degrees is hot enough to burn you real bad, but not hot enough to ignite paper, much less harm metal. Vintage bargain basement amps had trannies that ran hot enough to burn you. Core temp is the real problem, but basically unknowable. Insulation failure is the biggest threat. 2. I do believe a fan is an effective cooling solution.
-
I vote for a filament transformer and a fan.
There is no such thing as being to reliable. :icon_biggrin:
-
You can use an infrared thermometer. I got one for about $17 at Harbor Freight.
This is a rare occurrence where I disagree with Ed. 1. I don't believe a Tranny that's hot enough to burn you is necessarily a problem. 450 degrees is hot enough to burn you real bad, but not hot enough to ignite paper, much less harm metal. Vintage bargain basement amps had trannies that ran hot enough to burn you. Core temp is the real problem, but basically unknowable. Insulation failure is the biggest threat. 2. I do believe a fan is an effective cooling solution.
JJ, I do not see a disagreement at all. Fans INSTALLED in a head are really nice and I put them somewhere in every amp. I was not writing to you either as I know you know the difference between an amp that has a hot transformer caused by overcurrent and one getting hot from ambiant surroundings.
I have a few radios the transformers will blister you hands and play fine. Yep, they get hot. They live in Georgia and 2 of them always have, even when we did not have air conditining. Old Radios. Now if I were to hotrod one of these radios to run a power tube that took more current, I just took the hand of the Devil for a Dance. Didn't I just ask more from the transformer than the engineer of the radio had planned?
It is the core tempature, but aren't we speaking of an increase of 6 times over 6550 to KT120? This is a tube bass amp and with modern electronics I just do not see the engineers planning for the additional needed current.
So let me ask you directly. Without you checking the amp are you absolutely certain this Power Transformer will run these tubes full throttle, outside in 95 degrees with a fan blowing 95 degree air on the transformer and it will hold up without a problem? The OP certainly will push this amp hard.
Well I guess if you put it that way. :laugh:
-
Well a Leslie 147 Iron has plenty, but if you install 120's in it and get the correct cathode resistor to get close to 90% dissipation the voltage drop is too much.
90% current draw at idol in AB PP?
With a Leslie (and most amps?) wouldn't you want a more conservative/cleaner 70% at idol in AB PP?
-
Well a Leslie 147 Iron has plenty, but if you install 120's in it and get the correct cathode resistor to get close to 90% dissipation the voltage drop is too much.
90% current draw at idol in AB PP?
With a Leslie (and most amps?) wouldn't you want a more conservative/cleaner 70% at idol in AB PP?
Yes, if you maintain the UL setup, but a Leslie sooulds much better with Guitar if you Cathode Bias. At least it does to me. The heavy lows leslie cabs can make will stick around. I should have said this is cathode bias.
I do not know anyone who uses a Leslie in the guitar rig and maintains the stock output section. Sort of want it to blend so some headroom has to go.
-
Ahh, ok. Thanks Ed. :icon_biggrin: