Cornford hurricane Blown Mains TX 2nd time

[blindboybenton]

Hi Bob

The amp came in with a blown tx. when replaced i checked the tubes ( Sovtek ). they were ok but not matched anymore 2yrs old . so a new set of sovtek ( Matched ) went in
The tad bias master says at idle they are 30ma ea. I did think its a little high

Im re testing it now Full on output tubes are running at 48ma and 52ma When reduced to just before the sinewave clips 30ma ea
at full power Vk 12.9v rises to 22v. and pin 9 el84 338v under full load 264v
During this test 130 ma draw is max

Im going to adjust the bias and see what the current draw is then. Ill keep you all posted. Then i can get rid of the thing and get my loan amp back

BBB

[blindboybenton]

I have a aldi at the bottom of my road. I shall go when it stops raining.. Thought it was summer... Good old germans

[Enzo]

Here, this seems like a reasonable article on the subject.

Light Bulb Limiter
Doktor Ross Sewage: Tools of the Trade: Light Bulb Limiter

Using the bulb prevents fuses blowing and reduces current through the circuit when there is a short or other problem.

[blindboybenton]

Here, this seems like a reasonable article on the subject.

Light Bulb Limiter
Doktor Ross Sewage: Tools of the Trade: Light Bulb Limiter

Using the bulb prevents fuses blowing and reduces current through the circuit when there is a short or other problem.

Many thanks Enzo.. I will read it later. Ive been told of a place where i can get the bulbs too

BBB

[bob p]

The glass is probably made from orphans' tears to make you feel even more guilty about buying them.

LOL!

[Mick Bailey]

A bulb limiter gives a better indication of transient spikes in current which are too fast for a digital meter to respond to (unless you have a logging meter and set a audio warning threshold). It can also be positioned at your most sensitive angle of view - slightly off to your main line of vision - while you're probing or whatever. That's why drag racers don't look at the light, our ancient brains are wired to detect danger in our peripheral vision with more sensitivity.

A 130ma transformer should be more than enough for this amp. The 18w Marshalls were 120ma. Drawing 140ma is too much, but if you're driving the amp hard with a signal that's not representative of a guitar's signal you may get unexpected results. I recently worked on a Bird Golden Eagle amp that would suddenly pull excessive screen current when driven to a certain point with a signal generator. Also, a purely resistive load doesn't give a true representation of what current would actually be drawn with different frequencies into a reactive speaker load that's interacting with the OPT.

Another reason for excess current draw could be that the amp is getting pushed into ultrasonic oscillation.

[drewl]

Is the primary winding opening or one of the secondaries?
Check your filament wiring and the pilot light wiring, I had an amp that had shorted pilot light wiring which was tapped off the filament string that killed the power tans.

[bob p]

Doktor Ross Sewage: Tools of the Trade: Light Bulb Limiter

that fellow uses some rather "picturesque" speech in his narrative.

[bob p]

I agree with Mick, but looking the numbers, something doesn't add up about those test conditions.

Correct me if my understanding is wrong on this, but I think BBB mentioned that under load conditions, the amp is drawing 140mA on the B+ line. but if you do the math, the output tube loop is only drawing 100mA of cathode current. that leaves 40mA of current that's not accounted for. now you have to figure out where that 40mA current sink is hiding.

the amp has a compliment of 5 x 12AX7, and they're rated for a maximum plate dissipation of what, 1.2 mA each? 1.2 mA x 5 = 6.0 mA. But the B+ line is drawing 40mA.

I mentioned the need to draw-up a complete tube chart. The idea keeps getting ignored while you focus on the output tubes instead of looking at the big picture. I wasn't kidding about the need to draw up a complete voltage chart. There's ~40mA of B+ current outside of the power stage loop that still needs to be accounted for. You can't continue to ignore it. It's probably the reason for the recurring failures. It won't go away by turning a blind eye to it.

I'm thinking it's time to draw a complete tube chart and start calculating current in Kirchhoff loops.

[Mick Bailey]

The figures behind the static conditions are a little puzzling; "Amp on with No signal 79ma ( EL84) in place ... With El84 removed 9.5ma". So, subtracting 9.5ma from 79ma gives 69.5ma. Vk=12.9v. As the total current through the tubes has to flow through Rk the total current draw for the tubes is 58.6ma. So there's an additional 10.9ma current draw when the tubes are inserted that isn't passing through the tubes. Or is my reasoning flawed?

Another thing to check - If pin 9 is getting pulled down from 338v to 264v under full load, measure the voltage drop across R1 to give the total screen current.

[blindboybenton]

Many thanks to you all

Ill check it all again and give a full voltage reading of all tubes in and out + at idle and full power
Ill also give the current draw though heaters and 295v ac

What should this amp draw on full power ? I have some old Brimar EL84 and the original Sovtek el84. ill try with them

To answer the question by Drewl The secondary burns up.

Many thanks again

BBB

[Mick Bailey]

The power required at full output depends on the efficiency of the output stage in relation to the amp's rated power output. That is to say, if the amp produces 20w and the efficiency of the output stage is 50%, then you'd need 40w of power going in to that stage (not including transformer and other losses). The Hurricane is a class AB amp. The maximum efficiency for class A is 50% and class B 78.5%. AB lies somewhere between, depending on biasing - it can be biased more towards class A or towards class B.

If you look at PT transformer sizes for a typical 2xEL84 amp you'll get the idea that they're sized between 1.5x and 2x the quiescent current draw of the output stage. If you're biased at 30ma per tube, total is 60ma, x2 = 120ma. Coincidentally how the Marshall 18w is sized. 3x would be quite generous.

I think EL84 amps are the most dubious when it comes to output power claims. You see 12w all the way to 25w. Most will put out 15w continuous.

[J M Fahey]

To answer the question by Drewl The secondary burns up.

Thanks, but ... *which* secondary burns up?

And what does it mean "burns up"?

What is the symptom?

Opens? Shorts?

Supposing it shorts, how do you know *that* secondary is shorted and not some other?

Thanks again.

EDIT:
rereading the thread found a couple extra details:
1) in the picture I see the Tx you are using for testing is beefier (thicker EI stack for one) than original.
You also state:

The tx im using for test is the TEC323A
Using this tx im having no troubles with the amp Its just ive got the replacement one and its only rated at 130ma.

I think both the original TX and the replacement are marginally rated , and the test one is fine.
Just fit it there and sleep like a baby.
I'm sure the owner won't balk at it, nor to paying a little extra for an improved part.
Why insist on something already proven weak?
Of course, test the amp thoroughly before sending it back.
What I use (no kidding, I'm serious) is to play a few AC DC's albums, full power, sound horribly distorted (somehow AC DC is bearable under those conditions, that's why I use it) into a real speaker (no resistive load).
You'll need a speaker which can take it and if too unbearable, point it towards a wall or better into a clothes closet.
In short, reproduce what it will suffer at a rehearsal room, 4 or 5 hours in a row, balls to the wall.
You must be meaner than the owner himself.
Do something else meanwhile but every 30 minutes check the transformer is not bubbling/smoking/etc. , that tubes are not redplating, etc.

FWIW I had to move from a posh neighbourhood (Recoleta) to Buenos Aires's old port (La Boca, I'm barely half a block from the riverline and anchored ships) to be able to make such "live" tests.

2) don't worry about differences such as:

BBB mentioned that under load conditions, the amp is drawing 140mA on the B+ line. but if you do the math, the output tube loop is only drawing 100mA of cathode current. that leaves 40mA of current that's not accounted for. now you have to figure out where that 40mA current sink is hiding.

because they are not the same mA.

The "140" are read on an AC mA scale, and the current waveform is weird narrow pulses, what happens when you charge a capacitor with 50 Hz mains through diodes
What the display shows is an integration of that, with unknown time constants, chosen by the meter designer.

The "100" is read on a DC scale

Although in theory a "time constant" would be unneeded here (since the meters expects to read a static value), in fact there is one: in a needle meter you have the inertia of the mechanical parts, in a digital one there is a time constant introduced so the display does not continuously flicker.

Both measures will probably match if reading sinewaves, but here we have 50Hz narrow charging pulses at the HV AC side, and audio frequency halfwaves at the cathode resistor ... which to boot has a capacitor across it.

In a nutshell, I'd be *very* surprised (even worried) if both values matched !!!
So 100 and 140 are in the ballpark and close enough for me.

As of:

I know the customer wont run it flat out.

or

Accept that customer is not going to run a full power.

I wouldn't count on that, quite the contrary.

those are test conditions and the customer is never going to run 1 kHz sine waves through the amp,

Agree. He's going to do much worse.
Be ready for that.
(remember the AC DC hint: overdrive, highs, lows, all together)

[blindboybenton]

Thanks, but ... *which* secondary burns up?

And what does it mean "burns up"?

What is the symptom?

Opens? Shorts?

Supposing it shorts, how do you know *that* secondary is shorted and not some other?

The maker didnt tell me. All i know is the mains fuse ( 500ma ) Blown when you replace the tx would smoke then blow it again.

I will do some more checks later on and see what i get

BBB

[blindboybenton]

The power required at full output depends on the efficiency of the output stage in relation to the amp's rated power output. That is to say, if the amp produces 20w and the efficiency of the output stage is 50%, then you'd need 40w of power going in to that stage (not including transformer and other losses). The Hurricane is a class AB amp. The maximum efficiency for class A is 50% and class B 78.5%. AB lies somewhere between, depending on biasing - it can be biased more towards class A or towards class B.

If you look at PT transformer sizes for a typical 2xEL84 amp you'll get the idea that they're sized between 1.5x and 2x the quiescent current draw of the output stage. If you're biased at 30ma per tube, total is 60ma, x2 = 120ma. Coincidentally how the Marshall 18w is sized. 3x would be quite generous.

I think EL84 amps are the most dubious when it comes to output power claims. You see 12w all the way to 25w. Most will put out 15w continuous.

Many thanks Mick

To be fair this is the first EL84 amp ive had troubles with. Ive only had easy jobs on this type of amp. tube changes or cap jobs. Put me off the damm things for life.

Going off subject abit

where are you from. Hows business ? I do it on the side at the present but seem to be getting busy every month.
I will hopefully go part time in my day job.. Add the amp repairs and bands + alittle Bass teaching im busy

BBB