HT-5R combo makes hum through speakers in standby

[dioman]

It seems to me hum is 60hz, When I turn off standby hum become quieter. Also it heats up after 30 minutes of playing to the point that I can't touch knobs. I believe it is serious since there should be no current on speaker in standby.

I bought it used and found that there is soldering work on R39, it is replaced with more powerful resistor, although R38 is not changed, I checked they both ~22K, so I guess it should be ok.

Any advise on how I can diagnose?

[thephantum]

There is a known issue with a small subset of HT-5's. The issue is a short on the PCB that causes R39 to burn out.

http://www.blackstaramps.com/forum/view ... =32&t=2135

***Do not continue to power on the amp.*** The fact that R39 was replaced with a more powerful resistor coupled with the fact that the amp gets too hot to touch tells me the short was not fixed. The amp is now a fire hazard. Either get it to a tech and have it repaired correctly or repair it correctly yourself.

[dioman]

There is no short near R39, legs of capacitor and transistor are soldered with R39 in the air, looks dirty but completely fixes short and correct - I checked with schematic, also checked resistance to make sure that fix is working.

I want to fix myself, The fact that speaker hums in standby - maybe there is another short somewhere

[thephantum]

If the amp is getting too hot to touch, it is either not fixed correctly or there is another short that needs to be found. Either way, the amp is a fire hazard, so proceed with caution. There is somewhere around 320V in that area of the PCB and it was obviously a very amateur repair job, so that is where I would start.

The original short was between two pads of TR3, but I'm not sure which two. It's possible that it is between the other two legs of TR3 which is causing R39 to get too much current. The only way to find out for sure is to put the board under a microscope and look for signs of arcing.

Another approach would be this: If R39 is anything other than a 22K, 1W resistor, put the correct value resistor back in and see what happens. If it burns out, that will answer the question if the short was fixed or not.

[dioman]

Another approach would be this: If R39 is anything other than a 22K, 1W resistor, put the correct value resistor back in and see what happens. If it burns out, that will answer the question if the short was fixed or not.

Did this, also replaced blown capacitor (only thing that appears to be broken from visual inspection), now no heating, no hum, but it also doesn't work - dead silent, only thing is I hear cracking sound when strumming on max volume/gain. Looks like I won't be able to fix it myself

[dioman]

At last I received my new PCB - replaced it. And it's working. It sounded horrible at first, I biased it as advised here:
http://www.tonegeek.com/musicgear/amps/ ... -sound.php

Set R33 voltage to 46mV, sounds excellent to my ears

2thephantum: is it good setup? Or it need to do something else

[thephantum]

I wouldn't set bias that way. It's way too cool...but if the amp sounds good it should be fine. The tube will actually live a little longer.

Here's why I wouldn't set it that way: The HT-5 is push-pull, Class AB amp. That seems counter-intuitive since push pull requires two power tubes and the HT-5 only has one. However, the 12BH7 is a dual triode tube. So the amp uses both "sides" of that tube in a push pull config. You bias these amps just like any other push-pull class AB amp.

So what does all that mean?

If I use the numbers from that link: he had a Plate voltage of 470V. Based on the plate dissipation of the 12BH7 (3.5W) and going with the defacto standard of 70% max plate dissapation, that gives me a bias value of 5.22mA when measured directly at the cathode. I would measure that across 1 ohm resistors installed between each cathode and ground to give me that value in mV (Ohms law). What he did is measure it across an existing common 4.7 ohm resistor (R33). So, using Ohms law, that gives me 24.5mV per side. Since R33 is installed at both cathodes (common cathode), I need to double that which gives me 49mV.

So, let's look at the drawbacks of that method:

• - 45mV is a little colder than the standard 70% of max plate dissipation.
  - The HT-5 sounds better when biased closer to 80% max dissipation, which is closer to 55mV.
  - All of that is assuming a plate voltage of 470V. That measurement varys depending on the amp and tube used. My own HT-5 measured 453V at the last tube change...that's a pretty big difference...
  - It does nothing to address balancing the bias current across both triodes. Even a balanced tube can have a difference of 5%. That means one side could be running at 23.25mV and the other at 25.75mV. That's a 2.5mV swing. The amp will perform better if they are the same.

The point is that there is not "one size fits all" bias setting. Every amp is different, every tube is different. You need to measure, do the math and adjust. Then you need to do it again, because when you adjust bias, the plate voltages change (Ohms law again). So you repeat the process until the amp is dialed in. You should also measure and adjust at each cathode....not at a point common to both.

My $0.02

[dioman]

I set bias and balance and measures are (with ground),

Pin1 438 V
Pin2 22.5 V
Pin3 56 mV
Pin6 438 V
Pin7 22.6 V
Pin8 56 mV

So looks like it is exactly 70%(5.6) for this plate voltage ... it sounds awesome, but maybe too gainy, starting to break after half volume on clean

Do you think it's too hot?

[thephantum]

It's probably not hot enough for my tastes.

The HT-5 does not have a lot of headroom, that's actually one of the things I like about it. To my ears, overdriving the clean channel sounds better than the gain channel does. Running the tube cooler wont improve the headroom at all...it just makes the amp sound flat and lifeless.