| Vespasian |
After being powered up for an hour or so, how hot should an output transformer get? Can't keep my palm down on my early 60's Scott for more than 3 or 4 seconds.
Thanks again, Ed |
|
|
| Wavebourn |
| quote: | Originally posted by Vespasian
After being powered up for an hour or so, how hot should an output transformer get? Can't keep my palm down on my early 60's Scott for more than 3 or 4 seconds.
Thanks again, Ed |
Usually power transformers are calculated by overheating or voltage loss. So, it is usual for them to be hot. It may be too hot if designed for 115V when you have 130V, it may mean a core saturation. Also, your output tubes may be biased too heavy.
I don't know if it is the case, I hope I might help. |
|
|
| Vespasian |
So... based on the 3 or 4 second statement above, does that sound too hot to you? I've got 122v here... Thanks for your input Wavebourn.
Cheers, Ed |
|
|
| jaymanaa |
| Most of mine get that hot, at least on the store bought stuff. |
|
|
| sorenj07 |
| a quick fix is to slap a thermistor or two on the transformer's primary winding. dual benefit: slow start, and a bit of a dropped voltage :) |
|
|
| DigitalJunkie |
Power trannies usually run warm/hot..
Output Transformers,IME,Don't usually run HOT,or even very warm... you might want to check the bias and whatnot,and make sure the amp isn't oscillating. |
|
|
| HollowState |
Hi Ed,
You did say and mean output transformer? I concur with DigitalJunkie. An output transformer should not run very hot, even after many hours of operation. Most of their heat comes from what is absorbed from the nearby tubes and power transformer. If you can't keep your hand on it for more then a few seconds, it's up around 125º F or more. And that's too hot. Do scope the output for ultrasonic oscillation.
Victor |
|
|
| refference |
Hi Vespasian ,
OUTPUT TRANSFORMER don’t run hot , by itself .
You need to check at least 4 points :
1) If the power amplifier is oscillating ( as said above )
Use a scope .
2) If you have a biasing problem ( as said above )
Check the values
3) Check the negative feedback network ( capacitors ,
resistors , etc.)
4) You probably have a defective output transformer
( a short circuit between two or more turns , in the
primary and / or in the secondary windings )
Two months ago I had a similar problem with my client’s
SCOTT 240 and the solution , was to rewire the output trans-
former .
I hope it helps the troubleshooting . Regards ,
Carlos |
|
|
| Wavebourn |
| Oops... I misread, it is about output transformer. Too much bass too loud may cause the similar effect. |
|
|
| Vespasian |
Thanks everyone, - both output transformers are running hot. The amp is using 7591's. Could someone give me a step by step on how to properly set the bias? There are two pots for dc bias, and I simpy have them set so the voltage difference between the plates is 0v. The plates are at about 435 and grids are around -17, -18v. Don't have a scope...
 |
|
|
| Tube_Dude |
| quote: | Originally posted by Vespasian
Could someone give me a step by step on how to properly set the bias? There are two pots for dc bias, and I simpy have them set so the voltage difference between the plates is 0v. |
Two pots per channel or one for each channel? |
|
|
| jaymanaa |
| quote: | Originally posted by jaymanaa
Most of mine get that hot, at least on the store bought stuff. |
I mis-read as well. Thought it was PT. Ooops. |
|
|
| Frank Berry |
The output transformers on my HH Scott model 299C run cool.
The two pots are bias balance pots. Adjust them for a plate-to-plate reading of 0 volts.
The selenium bias rectifier bridge should be replaced with a silicon bridge. The old rectifier bridge produces a lower bias voltage as it ages. This could be the cause of your hot output transformers. |
|
|
| bembel |
I agree with DigitalJunkie, HollowState & refference.
It shouldn't be that hot ! (even with heavy bias mismatch problems, I never saw such a hot OT). |
|
|
| Vespasian |
Yes, one pot per channel. A new rectifier is installed and providing -46.3, schematic calls for -45. All the coupling caps have been replaced.
Can someone describe "ultrasonic oscillation" - I am not familiar with what it is, how it is caused or what the rectification would be.
Are there ways to check the transformer's health via measuring resistance?
The amp is really sounding pretty good with the exception of a phase splitter tube that is causing some hum in one channel.
Bad transformers? That would be a real drag.
Cheers, Ed |
|
|
| tubelab.com |
I doubt that your transformers are bad. Even if there was a few shorted turns in the transformer you would have to play it very loud for the transformer to get hot. If this were the case, it wouldn't sound good. (not the case for a bad power transformer)
Most OPT's get hot by absorbing the radiant heat from the output tubes. I have never seen them get too hot to touch though. You can check for this by puting a shiny reflective surface between the output tubes and the transformer and running the amp for an hour. If the transformers are much cooler, you can quit worrying. I have used a small mirror or aluminum foil for this test. Obviously, make sure the "shield" is not touching anything other than the chassis. The metallic surface on a mirror IS conductive!
Are the end bells getting hot, or is the whole core (laminations) getting hot? It takes a lot more energy to heat up the whole transformer core.
On amps that are crowded and get hot (including some that I have built) I often add a small fan. I use the fans that were made for computer chips. They are made to work on 12 volts. If you use a diode and a cap (1000 uF 16 volt) off of a 6 volt filament winding you get about 7 volts which will spin most of these fans enough to move some air, but not make much noise. Direct the fan at the output tubes. |
|
|
| Vespasian |
Thanks Tubelab - yeah the cores are getting hot. I'm hoping it is because I'm passing really low frequencies or it has something to do with the oscillation thing that was mentioned.
Cheers, Ed |
|
|
| Miles Prower |
| quote: | Originally posted by Vespasian
Can someone describe "ultrasonic oscillation" - I am not familiar with what it is, how it is caused or what the rectification would be.
|
It's an oscillation that generates frequencies too high to be heard. As for what causes this, it's inherent phase shift that turns intended negative feedback into positive feedback when open loop gain is still greater than unity. The best way to detect this is by o'scoping the outputs. If you don't have an o'scope, there are other ways to check for this. One thing to look out for is an amp that seems to go into clipping and overdrive prematurely. Something else to look for are finals that seem to be drawing too much current when biased to the design Q-Point Vgk. The excessive current means that there's a signal (the oscillation). Another thing to do is operate a radio receiver (preferrably a SW xcvr and check each ham band) but a BCB AM will do. If you hear noise when the amp is on that isn't there when it's off, then you can be pretty sure you have an oscillation problem.
| quote: | | Are there ways to check the transformer's health via measuring resistance? |
Only if you know what the design nominal coil resistances are.
| quote: | | The amp is really sounding pretty good with the exception of a phase splitter tube that is causing some hum in one channel. |
Could indicate an oscillation problem. Cathodyne type phase inverters are usually the first stage to overdrive since they have the least "headroom". The humming could very well be a heterodyne product from an RF oscillation if that stage is being overdriven into a nonlinear operation. |
|
|
| HollowState |
Ultrasonic is a term for a band of frequencies above the human hearing range. Typically from about 18KHz up to 50KHz or more. The quickest and easiest way to check for these is to put a scope across the amplifier's speaker terminals both with and without music.
What you do not want to see is a high level waveform without music, or overriding the music. You can't hear it but your tweeter can feel it and could burn out from too much of it. Try to borrow a scope if you don't have one.
Possible causes are internal generation from aged componets or altered wiring. It could also be passed along from the input. While this problem is rare, it's something that should be checked. A wideband AC voltmeter would also be useful.
Victor |
|
|
| Vespasian |
Thanks for the input guys.
The hum I mentioned that exists in one channel, does follow the tube (6gh8's) when I swap them.
I don't hear any clipping or distortion when I turn the amp up. Sounds pretty good actually.
I will try the test with a SW.
Not really sure how to check the current draw from the power tubes. I could use some help with that one.
I do appreciate all the assistance gotten on this forum!
Cheers, Ed |
|
|
| HollowState |
Hi Ed,
If this were my amplifier and I wanted to measure output tube current, here's what I would do.
I'd install a low value carbon comp resistor in each tube's cathode to ground. By low value I mean around 4.7 ohm (½ watt) or something close to that. I could then measure the small voltage drop across that resistor and use simple ohms law to calculate the current. (I=E/R or voltage divided by resistance)
For a current of 50mA, you would measure 0.235 volts across 4.7 ohms. And you could even make that cathode point available with a feed-through standoff on the chassis near the tube for future measurments. And do not be concerned about adding a tiny resistance in the cathode cirucit. Pragmatically it will have no effect on the sound or circuit performance.
If you'd prefer not to alter the amp in any way, then you could make an adaptor socket to place under the tube that would make the cathode and resistor available for your test probe. It they're octal base tubes it's easy. Just use a defective tube's base as your plug with a socket mounted on top.
victor |
|
|
| Vespasian |
| Thanks - Got it! |
|
|
| Vespasian |
The way this thing is laid out, the output tubes are right next to the transformers. On some other amps I see from the same period, they are not nearly as close, like Fisher x100's or even Scott's 299d. One opt has got the rectifier next to it also. The problem is compounded a bit also because the replacement JJ 7591's are larger than original stock (jj's on left) and can actually TOUCH the laminations on the transformers if tilted. Perhaps Tubelab is correct and this is simply a radiant heat thing. After 20 or so minutes the far sides of the tranformers are much cooler than those facing the tubes.
 |
|
|
| tubelab.com |
The tube current can be measured without adding any resistors.
With the amp off, and everything discharged (I would measure it after it has been off overnight) measure the DC resistance of each half of the OPT primary. Write down the numbers. Turn the amp on, with no signal applied, measure the voltage across the same primary halves. Take the measurements after the readings have stabilized, but before the transformers heat up, the resistance will change slightly. You can use ohms law to compute the plate current. The screen current can not be easily measured, but can be assumed to be 5 to 7 mA per tube.
It is normal for the dc resistances of each transformer half to be slightly different. It is also normal for the currents to be slightly different. Adjust the balance pot for minimum hum.
I set a black RF attenuator on top of the electrolytic cap on one of my SimpleSE boards, This puts it about 1/4 inch from ONE 6L6 tube. It was too hot to touch after about an hour. |
|
|
|
