Blackstar HT-5 Tube Swap

Discussion - HT-5 amplifiers.
User avatar
thephantum
Posts: 1160
Joined: Wed Jan 29, 2014 7:42 pm
Location: Virginia, United States

Thu Nov 20, 2014 11:44 pm

Sounds like it's biased way to hot and the tube is red-plating. I really don't understand why everyone is trying to reinvent the wheel on biasing these amps. Contrary to popular belief, it's just a fixed bias amp. You can therefore bias it using the standard cathode resistor method (or if you have really good DMM's, like Flukes, you can use the OT Shunt method).

The only trick is you have to go back and forth between each triode to get the tube balanced. Blackstar runs the power tube right on the edge of it's specs and runs it "hot". So even with just a little more bias on the hot side (even on one triode), it will red-plate.

***Remember, there are lethal currents flowing. If you are not comfortable doing this, take it to a tech. I will not be responsible if you kill or otherwise maim yourself ***

Read the plate voltages across Pin 8 (Cathode 1) and Pin 6 (Anode 1). That will be the difference in voltage across the two, which is the plate voltage for Triode 1. Then do the same across Pin 3 and Pin 1. That's the plate voltage for Triode 2. On mine they both read 450V (which is the maximum plate voltage a 12BH7A can handle).

Then read the cathode currents (at Pins 8 & at Pin 3) across 1Ohm, 1watt resistors, which will give you a mV reading. From the factory mine was set pretty hot and was out of balance. One side read 6.3mV the other read 7.1mV. I turned it down to 6.2mV and balanced it across both sides. It's still pretty hot (at 450 plate voltage, a good setting would be around 5.5mV), but it sounds damn good and I have no issues with valves burning out.

It's easiest to do this with a bias probe...but try and find one for a 12BH7A. I ended up making my own for about $20 in parts and an hour or so with the soldering iron.
Image

Plug the probe into the the amp, the tube into the probe and measure from the pins on the box. The "schematic" I quickly sketched before the build is below. Basically it inserts a 1ohm resistor between the cathodes on the amp and the tube for reading bias. It also inserts a a couple of resistors between the anode and cathode of each triode so that the plate voltages read in mV....you don't have to do that. I did because I'm lazy and didn't want to have to switch between ranges on my DMM's. :mrgreen:

All the resistors are in the box, so this rig is pulling high voltages & currents out of the amp and into the box. So even though it is a fairly simple build (a box, some wire, 2x 9 pin socket savers, 6 resistors, 6 banana jacks and some shrink tubing), this is not a build for someone who is unsure of what they are doing and/or does not know how to test a rig like this prior to using it. So to reiterate:

***Remember, there are lethal currents flowing. If you are not comfortable doing this, take it to a tech. I will not be responsible if you kill or otherwise maim yourself ***
Image

bhgtan
Posts: 4
Joined: Wed Nov 19, 2014 12:10 pm

Fri Nov 21, 2014 12:58 am

Thanks ... This is great info

Will give it a shot and revert the outcome ...

Thanks again ...

Cheers ... 8-)

bhgtan
Posts: 4
Joined: Wed Nov 19, 2014 12:10 pm

Fri Nov 21, 2014 4:27 am

Hi ...

This is what i have understood and translate into a better picture for all ....

Pls correct me if this totally wrong ....

Thanks again
Attachments
HT5 Power Bias.JPG
HT5 Power Bias.JPG (62.04 KiB) Viewed 22021 times

User avatar
thephantum
Posts: 1160
Joined: Wed Jan 29, 2014 7:42 pm
Location: Virginia, United States

Fri Nov 21, 2014 3:50 pm

In a nutshell, that's correct. However, you should have 450V at the plates...that doesn't mean you will. Mine actually measured 449V. I know that's pretty damn close but, the point is that in a Class AB push-pull amp (which the HT-5 is), the bias setting is actually based on the plate voltage read and the desired plate dissipation of the valve. Ideally (and this is arguable), you want to run the bottle at around 70% of max dissipation. So the math is:
  • ((Max dissipation wattage *.7) / Plate voltage) * 1000 = Bias current
(You use the 1 Ohm resistors so that you read it in mV instead of mA...Ohms law)

So in the case of the HT-5, max plate dissipation of a 12BH7A is 3.5w. Assuming you have 450v at the plate, the math is:
  • ((3.5w * .7) /450v) * 1000 = 5.444mA
That's the reading for each "side" of the bottle, which is what the BAL adjust pot is for...to balance the bias current across both halves. It's important because each half could have slightly different plate voltages as well. ;)

The only reason I went higher is because it was closer to what BS had the amp set at from the factory...and it sounds much better hotter. 6.2 is actually 80% of max dissipation (BS had it set closer to 85% :shock: ):
  • ((3.5w *.8) / 450v = 6.222mA
Like I said earlier, they are pushing that valve very close to the edge of it's operating window:
  • - The maximum plate voltage of a 12BH7A is 450v...and that's where BS run's it
    - The maximum plate dissipation is 3.5w and BS set it to almost 3w (3.5 * .85 = 2.975)
Typically, a valve run hotter will sound better, but it won't live as long. Go too hot and it tips over...it can red-plate or just out and out fail. I've been running mine, with the stock valve at 80%, for over a year now and it's still going strong. I actually just acquired a bunch of 12BH7A bottles from an estate sale. Got them cheap and 4 of them tested through the roof. They are from the 1950's and 60's and I'm getting ready to roll them across my HT-5...which is why the probe was out on my bench for the picture. :mrgreen:

bhgtan
Posts: 4
Joined: Wed Nov 19, 2014 12:10 pm

Fri Nov 21, 2014 4:58 pm

Thanks for the confirmation ....

You got some really nice coveted bottles there ...

Have a great weekend ...

Cheers

theDogger
Posts: 4
Joined: Wed Dec 18, 2013 10:26 pm

Tue Jan 06, 2015 9:23 pm

Ok lets get some images up here of the trim pots...
Image
Image
Image

Now doing some research here is what I have surmised!

The HT5 amps are adjustable bias and there is a balance control as well to match each section of the 12BH7.

On the HT5’s I’ve have, I simply measure the plate current by referencing the voltage drop on the OT primaries or with a custom built bias probe and shown above. I did not notice the cathode being held above ground potential (R33).

The R33 is a 10Ω resistor. The max plate dissipation of a 12BH7 is 3.5 watts per side. So at 433VDC, divided into 3.5W would be 0.008V, reduced by 70% would be 0.0056V or 5.6mV. When you factor in the 10Ω resistor that moves the decimal place and puts you at 56mV. Mathematically BS 46mV recommendation is a bit conservative based on 70% max dissipation at idle. However 433VDC on the plates of a 12BH7 is the absolute design maximum so I can see why they’re a bit conservative.

*(On a side note if you want to know the literal plate current for each triode simply measure the DCR of each primary leg on the OT (with the amp off and discharged) and then measure the voltage drop across it. Ohms law from their gets you dialed in. This takes all the mystery out.)

As far as 12BH7’s go, I don’t care for any of the new production offerings. I would stay NOS for this. I prefer JJ tubes and JJ does not a make a straight 12BH7. They make a proprietary driver tube called an ECC99 which has the same pinout and is reasonably comparable. Unfortunately it’s a 5 watt tube (per side) and won’t bias up properly without modifications (I’ve played around with this and they don’t sound particularly good running cold).

So I am running a JJ ECC83S (12AX7) Matched & Balanced but sometime swap out for a ECC83S High Gain & Current Balanced.
theDogger

Tober
Posts: 4
Joined: Thu Dec 12, 2013 3:36 pm

Sun Feb 01, 2015 3:10 pm

Do I need to plug in a speaker while I try to adjust the bias? Do I leave the amp with the power on and in stand by mode?

I got a used ht-5 and sounded ok for a while but now my clean channel is breaking up really bad and is almost unusable. I ordered new tubes ( figured it couldn't hurt) but then found it needs to be biased. I have the ht-1 and that is a fixed bias, I thought the same would go with the the ht-5.

User avatar
thephantum
Posts: 1160
Joined: Wed Jan 29, 2014 7:42 pm
Location: Virginia, United States

Thu Feb 05, 2015 4:48 pm

It's actually the other way around...the HT-1 is cathode biased, meaning it does not have to be adjusted. Fixed bias means it does have to be adjusted (you manually "fix" bias in place).

The amp should be powered on with a speaker plugged in. Standby should be off (as if you are going to play) with the channel volumes at zero.

vik62
Posts: 1
Joined: Mon Mar 09, 2015 5:16 pm

Mon Mar 09, 2015 6:05 pm

Hi, here my little experience with Blackstar HT5R.
After two years I found the sound in "gain" position too nasal and too rich of high frequences. So I decided to try to change the tubes, but there is no an authorized repairer nearby.
The original tubes are: 1x 12bh7 Electro Harmonix, 1x 12ax7 Sovtek.
So I take the risk to do it myself with two NOS tube that I found in an electronic country fair: 12bh7 Tungsram and 12ax7a RCA (expensive as the new tubes but as I read much better in quality).
Result: not really great differences with the 12bh7 Tungsram, but a significative one with the RCA 12bh7: rich sound, tone and frequencies. I like the blues sound and with my LP the sound is now amazing (but obviously it is always a 5 watt amp).
I also think it isn't necessary to regulate the bias, perhaps it is very difficult and the tubes are not surely more hot than the previous two, what do you think about?
Vik

User avatar
thephantum
Posts: 1160
Joined: Wed Jan 29, 2014 7:42 pm
Location: Virginia, United States

Wed Mar 11, 2015 12:20 am

Bias should be at least checked every time the power tube is changed and then adjusted, if necessary. In my own experience, NOS tubes tend to run cooler (bias wise) than the stock EH tube, so you'll probably be OK. The key word there being "probably".

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 9 guests