• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Mercury Magnetics OPT

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi everyone, I'm replacing an output transformer in a Leslie model 25 amp. It's probably not the most familiar tube amp is history, but to sum it up, it's basically 12AU7 phase inverter and a class AB pair of 6L6 tubes which are cathode-biased. The customer wanted the output transformer replaced with the Mercury magnetics LES-25-O After installing it I immediately noticed the output tubes were biasing extremely hot (about 75 mA each! measured with bias probe) Needless to say, I shut it down and began retracing my work and measuring to see what may be causing this. I found that the mercury output transformer's primary is measuring about 500 ohms, whereas the original measures in at about 3.9K ohms (which is about what I would expect). Not sure if this DC resistance reading is related to the hot tubes, but definitely doesn't seem right.

I called mercury, but they were not much help. They kept telling me that DC resistance is different than impedance (which I am fully aware of), and also that any measurement I got from an original transformer is unreliable because their tolerances were all "over the place back then" Maybe so, but 500 vs. 4000 seems too far off to be tolerance related... Mercury could not tell me what DC resistance is that I should be measuring from their primary... When I asked what the number of turns and wire gauge was (So I could do an approximate calculation it myself), they told me that was information that could not be revealed... The only option they gave me was to ship it back to them so that they can check it out.

Feeling a bit frustrated, and pretty underwhelmed with help from Mercury, especially given the price of their stuff...

If anybody could shed some knowledge on the subject, I would greatly appreciate it. On average a cathode biased 6L6 should idle around 35 mA, right?
 
Ex-Moderator
Joined 2011
500 seems pretty reasonable (if a bit on the high side), but not sure about the original reading 3.9k though. In any case, swapping the OPT should not cause the bias to change that much, put the original OPT back and see if the bias return to their normal reading?
 
I think you have to supply some more details of the 6L6 tube circuit. What is the supply voltage to the plate (via the output transformer)? What is the DC voltage on the screen grid? The screen grid voltage (together with the transconductance gm) defines the control grids maximum negative voltage before cutoff. And perhaps the control grids negative biasing voltage is about half that cutoff voltage. This, together with the DC resistance of the primary transformer coil, determines the working point. Moreover 3k9 resistance of the primary part: isn't that perhaps 390 Ohms? jazbo8 gave a good advice to switch back to the original situation and do some biasing measurements then.
 
That's a good point. I guess everything is operating as it should?
But doesn't 75 mA seem excessive for a 6L6? Surely the lifespan of the tube will be considerably shorter...
Or is this a normal region for cathode biased amps?
I'm not sure. I'm am more accustom to fixed bias amps with about 35-40 mA per 6L6.
 
That's a good point. I guess everything is operating as it should?
But doesn't 75 mA seem excessive for a 6L6? Surely the lifespan of the tube will be considerably shorter...
Or is this a normal region for cathode biased amps?
I'm not sure. I'm am more accustom to fixed bias amps with about 35-40 mA per 6L6.

As others have said it is not likely the DC resistance of the original trafo was supposed to be 3.9K - that sure does sound like an impedance figure to me.

Also - at 75 ma and (410-36) volts across the tube it is dissipating just a hair over 28 watts. That's within spec for a 6L6GC but it is higher than most applications seem to run.
 
You have more than 410V - 36V = 374V on the screen. The screen is connected directly to B+(there is no DC drop in the output transformer, the screen is connected to the center tap). Be Sure to use 6L6GC tubes.
6L6 screen 270V max
6L6 screen 270V max
6L6GB screen 300V max

You might have one weak 6L6GC, and the other is hogging all the current (not easy to tell with the common cathode self bias resistor). Use 1 Ohm resistors in series with each cathode, 75 mA will give a 75mV drop (check with your DMM).

A medium amount of unmatched current will cause early saturation of the OT on Bass frequencies.
A large amount of unmatched current will cause one tube to run extremely hot (and the transformer will already be saturated even without signal).

Be sure to look at the voltage on the 6L6GC control grids. If they are not near 0 Volts it means one of two things:
1. Leaky 0.1uf (aged) coupling caps. If you replace, be sure to use 600V or 630V rated caps.
2. Gassy 6L6GC tube(s). Replace with Matched tubes.
 

PRR

Member
Joined 2003
Paid Member
... ... ... On average a cathode biased 6L6 should idle around 35 mA, right?

No. Almost never. Refer to 6L6 datasheet.

All the classic 6L6 recipes show more than 35mA. They only go to 19W Pdiss except an extreme case added for 6L6GC. Turn to 7027 data. This is nearly the same tube but always rated for higher voltage than original 6L6, and has specific cathode-bias suggestions.

Or refer to the schematic which you linked: 35V across a 250r common cathode resistor. 144mA total cathode current! Say 72mA Ik each side. Taking 5% Ig2 for 6L6, this is 68mA each _plate_.

... ... ... the original measures in at about 3.9K ohms.... Not sure if this DC resistance reading is....

4K is similar to the nominal audio load impedance, right? If that were true, you would lose much or most of your precious audio power in OT losses. No, OT DCR is more often 5%-10% of nominal audio load. So for perhaps 5K loading we expect 250-500 Ohms DCR. In-sight of your MM windings.

If you use a DMM to measure OTs, short the un-used winding. The high inductance of big iron sometimes confuses the DMM's reading cycle. Shorting another winding reduces the inductance a lot.

Because OT DCR is normally "small", changing the OT should have almost no effect on tube heat. Something is wrong, and probably several somethings, but probably not the OTs.
 

Attachments

  • Leslie25.gif
    Leslie25.gif
    20.1 KB · Views: 196
  • 6L6-7027.gif
    6L6-7027.gif
    26.5 KB · Views: 207
I totally agree in your statements on the primary's 500 ohms DC resistance figure and also on it's neglectible impact on the tubes' bias. But I'd like to ask some more questions: Why does the OP's customer the OT want to be replaced? Was the amplifier malfunctioning previously? Did the OP measure these ominous 3k9 by himself, using a standard DMM?
If so, I suspect an open primary winding with burnt insulation, hence a carbon bridge between the open ends.
Best regards!
 
I need to make sure my old VOM with resistance readings is at my test bench (not somewhere else).

Suppose I jumper/short all the windings except the one I am measuring, and measure the DCR with a DMM, and I get 297 Ohms. Then I take all the jumpers off the unused windings, and measure with my VOM (and my eyeball estimate says it reads somewhere between 280 and 320 Ohms. Do I want extreme accuracy, or do I want a simple quick DCR measurement? Now, where are those darned jumper clips/shorts hiding?

Oh, I am measuring 3.9k of possibly burned out primary wire with carbon bridge across the open/burned out wire. How can I check that? Use more current than the DMM or VOM resistance scales use. Get a 70V power supply with the current limit set to 15mA, and measure the resultant voltage across the primary winding. If you get more than 0.015A x 3900 Ohms = 58.5V (if you get 70V) the winding is now open (I doubt the small carbon bridge can withstand the 0.88 Watts). You can use either a DMM or VOM to measure the voltage.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.