• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

1967 Ampeg B15NF Run Away

#41
Well...I have replaced the 100K 2W original carbon film resistor in the bias circuit with a new metal oxide resistor at same rating. I did mount the resistor 1 - 2 mm off of the board. So, we now have the new Sprague Atom bias cap and the new 100K resistor. I tested the amp and after 20 minutes of running the plate voltage stabilized at 490V on tube A and 488V on tube B. The bias reading stabilized at -48.8V. So it seems like the circuit stabilized and did not run away on me. I then connected my bias rite on the power tubes and fired it back up. After stabilizing it read 490V/ 60.5ma current on tube A and 488V/ 57.9ma on tube B. This meant that the power on tube A was at 29.6W and 27.8W on tube B. Using a the pair of 6L6GE from Groove Tubes, these are rated at 30W 500V maximum. So I am running them very high at almost maximum on A 98.6% and 92% on tube B. If I wanted to run these closer to 70% I should be targeting 21W.

The good news is that the circuit stabilized! So if I want to move the resistance of the bias resistor that is currently 47K ohm in a way that moves me toward the 70% target, what resistance would you try?
 
#43
So, what happened with your measured -79 volts at the bias supply output???

I don't think you changed anything performance-wise actually unless the bias supply drifts closer to zero (less negative).

What you need to do is to make the bias voltage more negative, R34 (47k/2W) is the bottom half of the bias voltage divider on the schematic I have (B-15N, 1968). Increasing its value to 56k or possibly 68k should get you to where you need to be.
 
#44
This is a great question and I believe the answer lies in when I took the reading that gave me the -79V at the bias. When I first turned this amp on after these recent changes, I measured -79 but that was upon start up. As I think about what could have been happening, I might have pulled the -79V reading when the amp was just on Stand By. Because of this discrepancy, I just ran a measurement over time. At start up, AND OFF OF STAND BY, I read -52.00V. From there, bias tracked as follows:
Time In Min
0.5 = -52.00V
1.0 = -51.52V
1.5 = -51.43V
2.0 = -51.41V
2.5 = -51.46V
3.0 = -51.57V
3.5 = -51.68V
4.0 = -51.75V
4.5 = -51.85
5.0 = -51.91V
5.5 = -52.00V
6.0 = -52.09V
6.5 = -52.21V
7.0 = -52.29V
7.5 = -52.36V
8.0 = -52.46V
8.5 = -52.55V
9.0 = -52.61V
9.5 = -52.68V
10.0 = - 52.75

I stopped taking readings after the 10 minutes. Line voltage was at 123.3V
 
#45
1967 B15NF Bias Cicuit Schematic.jpg


Here is the schematic from inside the head tray showing the bias circuit section on my amp. So if I change the 47K to a 56K (or even a 68K)I should see a more negative bias which will reduce current right?
 
#46
Gumps said:
View attachment 1057703

Here is the schematic from inside the head tray showing the bias circuit section on my amp. So if I change the 47K to a 56K (or even a 68K)I should see a more negative bias which will reduce current right?
Click to expand...
Yes. That is a different schematic than I referenced, there is an additional 56k resistor that is not present in the newer revs. (Possibly addressing this very issue)
 
#47
The 100K resistor that you measured as 130K would cause the bias voltage to be less negative (closer to zero) which would increase the idle cathode current. Not the direction that you want. You want the bias voltage to be more negative to decrease the cathode current.

When measuring the bias, the amp should be in playing mode, standby off. You can pull the power tubes, install the resistor and measure the bias voltage to ensure that it is what you want. Then reinsert the tubes, power up, and measure the bias voltage again.

When you powered up and first measured the bias, the amp was cold. Readings will drift until everything reaches a steady state. So the drifting what you measured is not unusual. Ampeg recommends that you power up the amp, put it in playing mode, standby off, for at least 20 minutes, then set the bias. Readings can continue to drift, even over an hour. Line voltage can drift slightly over this period as well. Whenever you set the bias, record the line voltage, it's a handy reference. The bias should be set without an instrument plugged in. I like to set the dials consistently, volume down all the way, bass and treble flat (at noon in this case); if the amp has them, any tone boosts such as ultra-hi and ultra-lo should be set to off, any gain boosts turned off.

It makes it easier to determine the required bias resistance if you remove the 47K bias resistor and tack in a pot. Set the bias by adjusting the pot. Then remove the pot and measure the resistance. This will tell you what resistance you will need. The schematic below is for a simple bias pot board. For the purpose of a tacked in pot you don't need R2, choose a linear pot around 60K-75K, or include R2 (36K or two 75K resistors in parallel) and use a 25K linear pot. Once the bias is set, the required resistance is the reading between pot terminals 2 and 3 plus the value of R2.

BTW, if you ever change the 100K 2W resistor again, go with a 3W or 4W resistor. The 2W can get quite hot and a larger resistor dissipates the heat better. Check the size first though, it has to fit on the board.

bias pot.jpg
 
#50
Paul....no worries! I appreciate your generosity.

Can I ask....as i adjust the bias now...is there any problem increasing the 47k resistor to a value above that of the 56k that is part of the bias circuit with the 47k?
 
#52
I think you will end up somewhere between 56k and 68k, you might need to parallel a couple resistors to get the value you need.
 
#53
upload_2017-2-18_12-52-20.png


Here are some test results that I obtained from altering the bias voltage on my B15NF. Conclusions are that all settings still produce plate voltages that seem high. Adding bias (more negative) moved to a 70% power target but at the expense of higher plate voltage. I need your advice...
1. Which way should I go here...lower current (which helps in moving away from max power dissipation of 30W) and higher volts (which slightly exceed a maximum plate voltage spec of 500V.) or go with higher current (creating close to max power dissipation) but reduces plate voltage below maximum?

2. I want to run RCA black plate 6L6GC's. How will these hold up to any of these condition choices?
 

Attachments

  • upload_2017-2-18_12-52-2.png
    upload_2017-2-18_12-52-2.png
    50.6 KB · Views: 38
#54
I would set the amp at whatever bias sounds the best. Try it with the different biases, push the amp hard and listen.

Keep in mind that if you run a higher plate current, you wear out the tubes faster. How much faster in your case, not sure. Given a choice, I prefer a more conservative bias setting.

Something about the vintage 6L6GC (not sure about all the current ones, some are fine), it can take a higher plate voltage than the data sheets specify. The designers knew this, a lot of amps were run that way. The data sheet specs are for a long service life.

If the higher plate voltage concerns you, buy a pair of 7581A's (www.thetubestore.com - Tung-Sol 7581A Audio Tubes). They have a 35W plate dissipation. They are a cost effective solution. I use them in place of hard to get NOS 7027A's, the perform very well. It's basically an enhanced 6L6GC.

The black plate RCA's will do fine at 70% max plate dissipation. Given how expensive these tubes are, I'd rather run the 7591A's.
 
Last edited:
#55
Plate voltage will naturally rise with a decrease in plate current. Measure the plate supply voltage with the amp in standby and it will be higher still.

Target for a plate power around 70% of maximum, that's probably the safest spot.

(edited for pre-coffee auto-correct)
 
Last edited:

Latest posts

Top Bottom
Back