diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Tubelab (http://www.diyaudio.com/forums/tubelab/)
-   -   TSE question.. (http://www.diyaudio.com/forums/tubelab/211366-tse-question.html)

Green77 23rd April 2012 07:27 AM

TSE question..
 
Hi all!

This weekend i was trying to troubleshot why i have some startup hum in my TSE.
It hum (almost as a groundloop) really loud for like 1 min and then it get lower. After 3 min it's dead silence.

I was using a scope and i was looking at the heater voltage. On the 5482 tube i was reading 5.5 V. A bit to low so
i put a 2.2 ohm resistor in parallel with R3 (2.2 ohm), then i have 6.25 v.
But my scope looked like this. Is that normal?

http://i43.tinypic.com/250nj9v.jpg

It looked the same before and after the 5 volt reg. too.

I don't think it has something to do with my startup hum because it still looks like this when the hum i gone.

Regards //Daniel

tomchr 23rd April 2012 05:35 PM

1 sec/div? Are you sure? With 50 Hz mains, you should get above waveform on a 2 ms/div scale. I'd expect the period to be 10 ms (1/(100 Hz)).

What you're seeing is ripple voltage. I'd pull the 2.2 ohm you added in parallel with R3 off and start debugging from there.

Are you having issues with the output voltage from the filament regulator, U1? If that's the case, first ensure that you have sufficient voltage going into the regulator. What's the lowest voltage on the waveform above? Measure with the scope DC coupled. The input voltage to the regulator IC needs to be at least 5 V plus the worst-case drop-out for the regulator (assuming 5 V filaments).

My guess is that your filament transformer is pushed beyond its limits. Or perhaps your mains voltage is running a bit low. What's the AC voltage of the filament winding under load?

During start-up, the filaments draw much higher current than they do after 30 seconds of warm-up. I'm guessing the higher current draw causes the voltage of the filament winding to sag. This pushes the ripple voltage below the minimum required input voltage for U1 resulting in hum on the filaments in the output tubes (which tubes are you using?) If you're using directly heated tubes (2A3, 300B, etc) this is guaranteed to cause hum. Once the filaments are hot, they don't draw as much current and the input voltage to U1 is now (barely) in spec. That's my guess. But without better measurements, it's hard to tell for sure. If you can, take a scope shot of the input voltage to U1 right at turn-on. Note the lowest voltage on the waveform.

Hope this helps.

~Tom

Green77 23rd April 2012 06:29 PM

Thanks! Sorry i didn't wrote that i use 300b tubes for output. I don't have the amp at home now so i can only tell what i remember :)

It looked the same on the scope before i added the 2.2 ohm resistor, only diffrent was that i only had 5.5 volt on the filment.
Iīm using a Hammond 372JX trafo (230 V primary). I have a 6,3V 8A heater winding. Input to U1 is 7 V when the amp is up and running.
I will check the heater with a scope tomorrow.

I will check the scope so i don't have pressed in any button, but the time switch was on 1 sec / div. And every cycle tock 5 div...

Many thanks for the help!

Regards //Daniel

tomchr 23rd April 2012 09:35 PM

Quote:

Originally Posted by Green77 (Post 2997872)
It looked the same on the scope before i added the 2.2 ohm resistor, only diffrent was that i only had 5.5 volt on the filment.
Iīm using a Hammond 372JX trafo (230 V primary). I have a 6,3V 8A heater winding. Input to U1 is 7 V when the amp is up and running.

Something does not make sense. First off, the 6.3 V, 8 A winding should provide more than enough power, hence, shouldn't sag under load.

If I interpret the schematic correctly, there's only one R3. So the filament current for both 5842's flow through R3. I'd use R3 = (7-6.3)/0.6 = 1.2 ohm (2 x 2.2 ohm in parallel would work as well). Note that you'll dissipate nearly half a Watt in R3, so use a 2 W type.

Quote:

Originally Posted by Green77 (Post 2997872)
I will check the heater with a scope tomorrow.

Check both before and after the regulator IC.
Before the IC you should see the waveform you posted in Post #1. The bottoms of the waveform should be comfortably above 5 V for the regulator to regulate properly. If not --> fix it (add voltage to the filament winding or increase C1). If the voltage doesn't drop below, say, 5.5 V on the input of U1 but you have ripple on the output of the regulator, the regulator is either dead or there's a solder blob somewhere near it.

Quote:

Originally Posted by Green77 (Post 2997872)
I will check the scope so i don't have pressed in any button, but the time switch was on 1 sec / div. And every cycle tock 5 div...

Are you using a sampling (digital) scope? If so, you're probably looking at an aliased signal. A 5 s period would be 0.2 Hz. Even if that could make it through the output transformer, you wouldn't be able to hear it. You'd be able to see the speaker cone move slowly back and forth, but that's it.

I bet that signal is 100 Hz (10 ms period).

Are you using the 5AR4 rectifier or are you using silicon?

~Tom

Green77 24th April 2012 06:20 AM

I'm using SS. I will try to look at it this afternoon. Thanks for the help!

Green77 4th May 2012 05:42 AM

Hi!

I finally have time to look at my TSE.
I scope the heater circut. After the 5 v reg and after the R3 (6,3V) itīs look like this.
And you was right tomchr! My bad, the scope was at 1 ms time/div. And on this pic 0,5V/div.

http://i45.tinypic.com/zmhbi0.jpg

But after like a minute the hum is gone and the TSE is dead silence and sounds greate!
I scoped B+ and it wasn't any ripple to talk about, 1-2 mv.
And the scope still looks like this... And the same after 1 hour too :)
So could it be a tube that hum when itīs not warm?

Thanks //Daniel

tomchr 4th May 2012 07:43 PM

Quote:

Originally Posted by Green77 (Post 3011179)
I scope the heater circut. After the 5 v reg and after the R3 (6,3V) itīs look like this.

You should not have that much ripple after the regulator (measured at C3). If you do have that much ripple, the regulator is not working properly. Check for soldering goofs, short circuits, and dead silicon.

I would expect that amount of ripple on C1 and R3. But the output of the regulator should be dead quiet. This tells me the regulator is not regulating. The most common cause of that is that the voltage drop across the regulator is not high enough for the regulator to regulate. Try to measure the voltage across the regulator IC. That regulator requires at least 500 mV across it to properly regulate.

The maximum input voltage for that regulator is 7 V, so if you've run it without load from a 6.3 V winding, you may have killed the regulator IC. Usually, there's a bit of margin from the ABS MAX specs to the actual point of damage, but I wouldn't rely on that for my designs.

Quote:

Originally Posted by Green77 (Post 3011179)
But after like a minute the hum is gone and the TSE is dead silence and sounds greate!
I scoped B+ and it wasn't any ripple to talk about, 1-2 mv.
And the scope still looks like this... And the same after 1 hour too :)
So could it be a tube that hum when itīs not warm?

Measure B+ as you turn on the amp. If you're using an analog meter use the 500 V or 1 kV scale. If you're using a digital meter, turn off auto range and set the range to the 500 V or 1 kV scale.

I've had amps hum a little when the B+ voltage was too low. If your B+ takes a minute to reach, say, 90 % of the target B+, the hum you're experiencing may be because the B+ charges too slowly.

What else happens when the hum disappears? Does this correlate with the time it takes the input tube to warm up? You can find out by measuring the voltage across the cathode resistor.

What does the grid voltage on the 300B do during warm-up? How about the 300B anode current (measure across R18)?

You say it takes a minute for the hum to disappear. Does it literally take 60 seconds? Or how long does it take for the hum to go away when the amp is turned on from being stone cold? What if you leave the amp on for a few hours. Turn it off for 15 seconds and then back on. Does the hum go away faster or slower?

You need to poke around a bit and provide some more data here... The amp should not hum on start-up. That said, my 6LU8 spud does hum for a second or two as it comes to life. That's because the B+ takes a little bit to build up. The hum is audible but certainly not loud. My 300B amp (very similar in design to the TSE) does not hum on start-up.

~Tom

Green77 4th May 2012 08:43 PM

I have looked at B+ when i start up the amp. It goes right up to 365 V (i use SS).
Bias looks the same (-225 V).

I will take the exact time on this tomorrow. It's night here and my second half is sleeping..She wouldn't be that impressed if i turn on some music now..i think :)

Thanks for the help. //Daniel

Green77 5th May 2012 11:38 AM

Hi.

When the amp is cold the hum starts after 2 sec. After 40 sec itīs decrease and is gone after 65 sec.
I played for 2 h and then switch of the amp and started it up again after 20 sec. It started to hum after 2 sec and dissapered after 7 sec..

So it's worst when its cold..

Thanks //Daniel

Alan Hope 5th May 2012 06:23 PM

Quote:

Originally Posted by Green77 (Post 3012537)
Hi.

When the amp is cold the hum starts after 2 sec. After 40 sec itīs decrease and is gone after 65 sec.
I played for 2 h and then switch of the amp and started it up again after 20 sec. It started to hum after 2 sec and dissapered after 7 sec..

So it's worst when its cold..

Thanks //Daniel

My TSE does pretty much the same - though the hum takes a bit longer to go away. Bidding on a 'scope right now on eBay so should be able to investigate this.

So watching this thread with interest ...


All times are GMT. The time now is 08:16 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright Đ1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2