5u4GB not working

[Bas van Dijk]

Hi All,

I'm new to this forum so let me first introduce myself.
My name is Bas van Dijk and I'm from Amsterdam, the Netherlands but live in Belgium.

I have a bit of experience in electronics but am new to tube amps. I've learned most from Uncle Doughs youtube channel.

The reason I'm looking into this is that I'm restoring the amp of a 50-ies RockOla Comet120.

I've started a blog, you can find some info on the amp here:

Amplifier | Rock-Ola 1438 Comet Fireball

No my problem is this: i've replaced all components in the amp but there was no sound.
I started testing and found there is no high voltage on the tubes.

I disconnected all connections to the 5U4GB rectifier, except the once that come from the power transformer.

I measure 5V AC between pins 2 and 8. The tube is glowing.
I measure 780V AC between pins 4 and 6 and 385 between either pin and ground.

But there is no DC voltage on pin 8 wrt ground.

Am I doing something wrong? Does the tube need a load to produce DC (I don't think so??).

The tube is new. am I just unlucky and got a broken tube? How can I test (I do own a oscilloscope btw).

Thanks, I'm sure I have many more questions as I go on, but many thanks for any info on this one.

BW,
Bas

 

[kevinkr]

Welcome to the forum! I lived in Brussels for a number of years as a teenager, many decades ago.

It does need to be connected to an input smoothing capacitor in order to make anything other than ripply looking DC that will probably confuse your meter.

You should be able to measure DC between pin 8 and ground (or pin 2 for that matter.)

I see from your measurements that the center tap is grounded.

Given that there is no smoke show or fireworks, and the voltages sound a bit high (i.e unloaded voltage) it doesn't seem likely that there is a short.

The only things I can think of is that the pins you think are pins 4 and 6 are not, and it's mis-wired (I have done things like this in the past) or the tube is somehow defective. Since it glows and is a comparatively simple device it is hard to imagine. Best approach would be to get another 5U4GTB to test it with.

Make sure meter is set to DC range when you go to measure the output voltage, I've done that too!

Others will weigh in soon, and hopefully will have some additional insights into the problem.

 

[Bas van Dijk]

Hi Kevin,

Many thanks for your reply.

Just curious: why do you say the voltage seem high?
The 5V is according to spec, and the 780 seems close to 50% of the 385V that is supposed to be the DC level so I thought that made sense for a half wave rectifier?

Always good to check but I'm sure that the meter is on the right setting

Thanks

Welcome to the forum! I lived in Brussels for a number of years as a teenager, many decades ago.

It does need to be connected to an input smoothing capacitor in order to make anything other than ripply looking DC that will probably confuse your meter.

You should be able to measure DC between pin 8 and ground (or pin 2 for that matter.)

I see from your measurements that the center tap is grounded.

Given that there is no smoke show or fireworks, and the voltages sound a bit high (i.e unloaded voltage) it doesn't seem likely that there is a short.

The only things I can think of is that the pins you think are pins 4 and 6 are not, and it's mis-wired (I have done things like this in the past) or the tube is somehow defective. Since it glows and is a comparatively simple device it is hard to imagine. Best approach would be to get another 5U4GTB to test it with.

Make sure meter is set to DC range when you go to measure the output voltage, I've done that too!

Others will weigh in soon, and hopefully will have some additional insights into the problem.

 

[Alllensoncanon]

Chances are the volt meter you have measure Vrms (root mean square) voltage. The circuit you use should produce DC voltages about 1.414 times of the Vrms AC voltage. Your measured 780V seems high. However, I suggest that you put back a minimum power supply circuit before start debugging. This will be as Kevinkr suggested , the first stage filter capacitor and a small load (bleeder resistor) of about 500K ohms. This ensure you are not looking outside of your amp's operational parameters (such as using a volt meter as a load and no filter capacitor). btw, you have a full wave rectifier circuit, just not using a bridged full wave rectifier circuit.

 

[Alllensoncanon]

In case the AC input voltage was truly 780V. You might want to find out what can happen to your volt meter if it measure something over 1000V, like 1103V. It could have burned out. Be care, you are working with fatal dangerous high voltages.

 

[PRR]

....the 780 seems close to 50% of the 385V that is supposed to be the DC level so I thought that made sense for a half wave rectifier?

No. This Rock-O-La is a Full Wave supply. Might be good to review rectifier types.

If you have 780V AC on the two rectifier plates, each plate has 390V to the center-tap, which is also circuit common.

390V Sine AC will peak to 546V. Without a first filter cap, you get a big bumpy wave which may not read right on some DC meters.

*With* a first filter cap, and no other load, the cap will charge to 546V DC.

Which means that the first 20uFd cap after the 5U4 "should" be rated for 546V! At least if you run it no-load for testing.

When the 6L6 tubes warm up, the total demand is over 100mA, apparently near 160mA.(*)

The PT and the 5U4 will sag. 5U4 data suggests it will sag to 440V. Rock-O-La may have used a 450V (common electrolytic) for the first filter cap, because these would stand some over-voltage for the one or two seconds between 5U4 warm-up and 6L6 warm-up.

(*)(Ah! This is a Field-Coil speaker. There is the 110mA in the 6L6es, and another 50mA to the speaker electromagnet. You better have that speaker in good shape or all voltages will be dangerously high.)

I do not know why your 5U4 appears to be not working. Possibilities are many. Bad 5U4. Bad socket contacts. Reading the wrong thing.

 

[Bas van Dijk]

Thanks for all the replies guys, Yes, indeed, I did forget about the AC vs peak so the voltage is high.

I will do some more checks tonight. The Voltmeter is still working (just checked).

I also do not think that the fact that there is no load is the issue (apart from the fact that it may give voltages that are too high) because I started with the whole amp connected and there was no sound, and then I measured no DC on any of the pins of my pre-amp tubes.

Thanks, great forum!

BTW I am being very careful!, but thanks for the warning Fully realize this is dangerous stuff.
I clip the meter on when the amp is powered down and switch things on hands off. And even then I am careful not to touch anything when it's disconnected in case any of the large caps are still charged.

 

[Bas van Dijk]

The 5u4 was broken! I replaced it and got an output. It does not sound good yet, but let's do that in another thread

Thanks