Shimmilou

Blues Junior Rehab

  I have owned and played my BJr for about 2-1/2 years. It is a great little amp, and I've enjoyed using it. I use it and several other amps, and I don't play that much to begin with, so it doesn't have a lot of miles on it. The BJr is one of the few amps that I have owned and played without first going through it for an inspection of the solder joints and capacitors, I just played it. I bought it to have a lightweight amp to carry around with me, and never got around to opening it up until I did a tone stack mod recently (added a jumper). While I had it apart I checked it over and the solder joints and caps were all in good shape. I saw the output tubes showed signs of overheating. The tubes are easy to pull and check out, so I pulled them to have a look. When I pulled the first output tube (EL84), the label, which used to be white and is now brown, just rubbed off on my fingers. On closer inspection I saw several small black spots of discoloration inside the tube glass (see pics at end of post). This is a bad sign, and both output tubes were in the same condition. The output tubes appeared to have been very hot. I decided that I would check the amp out before putting in new tubes. You may have read about the output tubes in the BJr running too hot, let's find out.

  First I took the following readings using a VOM, with the amp warmed up, before doing any modifications, with all knobs at 1 (zero). 330 volts power supply (B+), 326 volts on Plate, -10.86 volts on the grid bias (C-), 4 volts across P2 to P3 of the output transformer primary, at a measured 100 ohms resistance of that half of the output transformer primary (P2 to P3). To find out what the plate current is, and calculate the tube idle wattage we can use Ohm's Law. The plate current is the same current flowing through the output transformer primary. Ohm's Law says that, I = E / R, where I is plate current in amps, E is P2 to P3 volts, R is transformer resistance in ohms, solve for I. 4 volts divided by 100 0hms equals about 0.040A, or 40ma Plate current. For power, Ohms Law says that, P = I x E, where P is plate power in watts, I is plate current in amps, E is plate volts, solve for P. Multiply 326V (plate E) x 0.040A (plate I) = over 13 watts of plate idle dissipation. The tube data sheet shows a max plate dissipation of 12 watts for an EL84, and these tubes are idling at over 13 watts each!! The tubes have indeed been toasted because the bias was set much too high. Maybe they were trying to get a class A sound out of a class AB amp?? Maybe it was done to increase the sale of replacement tubes? Who knows why, but it certainly doesn't make the amp sound better, and can only lead to problems. Checking the tube bias chart for the EL84 with 330V on the plate, the plate current should be between 23.6 mA for class AB, and 29.8 mA for high AB. Even for class A amps with 330V on the plate, the plate current should be 34.5 mA, which is nearly 12 watts idle. The 40 mA existing plate current is much higher than the recommended range of 23.6 to 29.8 mA for class AB. If you go by the school of thought that idle plate dissipation should be 70 % of full power, then the tube should idle at about 8.5 watts. If you go by the bias chart the tube should idle somewhere between 7.8 watts and 9.8 watts, and 8.5 watts is right in that range. 7.8 watts idle would give more clean headroom sound, lower than that will start to sound harsh. Higher idle, up to 9.8 watts would start to give earlier breakup. With any more wattage than that, the sound just gets dull, and the tubes start to cook, shortening their life. With an idle of over 13 watts, I am surprised that the tubes lasted this long. Maybe it's a testament to Groove Tubes quality. ;)

  So, to start with, I knew that I needed to adjust the grid bias voltage, and since the bias is a set/fixed value in the BJr, I needed a potentiometer to replace the fixed value resistor in the grid bias power supply. It's possible to just calculate the value of the needed resistance for a new resistor in the bias supply. It would be much easier to change one resistor for another, but the bias voltage still would be fixed and not adjustable, so I went with the pot (you'll see why when I try different output tubes). The new 50k ohm pot has to be adjusted to the middle position before installing, to make sure that the starting point on the bias voltage isn't too low to avoid red-plating the tubes. The bias resistor that is being replaced is 22k Ohm and the middle position on the pot would give 25k ohms, which means that the bias voltage would already be better to start with, before any adjustment. While I had the amp apart to install the pot, I also installed 2 new tone stack caps, 2 new coupling caps, and added a cap to the first stage of the power supply. Adding the cap to the power supply doubled the capacitance value and added a few more volts to the plate, which is now at 330V. I also installed a standby switch by replacing the stock switch with a 3-position switch, which did nothing sonically, but using it allows less heat to be generated during breaks.

After installing the new parts and reassembling the amp, I rechecked all of the previous readings with the original output tubes installed. The new readings are; 333V B+, 330V on plate, -11.86 grid bias volts at C-, and 3.6V across P2 to P3. More negative voltage applied to the grid means less plate idle current, so the bias setting is already a step in the right direction. I adjusted the grid bias voltage to about -13V measured at C-, to achieve about 2.7V across P2 to P3, which means about 27mA of plate current. The recommended range of plate current is 22.9mA to 28.9mA, so 27mA is in that range. Using ohms law, inserting the new numbers measured, gives about 8.9 watts of idle dissipation, which is in the range given by the bias chart for this tube at this plate voltage. After experimenting with different plate currents and checking the sound, I set the grid bias voltage to just over -13V, giving just over 26 mA of plate current (about 8.5 watts idle plate dissipation). Everything checked out, time for a sound test with new output tubes!

  First, I put in a new set of 7189s (premium EL84s) in and switched the amp on and warmed it up. The tubestore.com says the 7189 is one of their "Preferred Series" tubes. Read about them here: 7189. With the new tubes installed, I checked the plate current again, and with no adjustment of the grid bias voltage, the new tubes were idling at about 28mA plate current, which is about 9.3 watts plate dissipation (note the different idle wattage of the new tubes at the same grid bias setting). I readjusted the bias to 26mA plate current, 8.6 watts idle plate dissipation. After warm up I played the guitar through the amp for a while, adjusting the knobs to explore the sound. When dialed in clean, the sound was very solid and very responsive to touch, with deep, tight low end, rich mids and crystal clear, shimmering highs and lots of harmonic overtones. The mods and new tubes made the amp sound much brighter, cleaner and clearer than when the amp was new. Then I tried some overdrive and found it to be even better sounding than before, smoother and louder with more sustain and still touch-sensitive, going easily from clean to breakup just by changing the pick aggressiveness. I thought that the amp sounded pretty good when new, but now it is a lot better, with more usable range on the volume. For the first time on this amp, I was able to turn both the volume and master volume to 12 and have a very good overdriven tone with lots of sustain and smooth, warm breakup, not harsh or muddy at all, and quite loud with very little noise or hum. With both master and volume at 12, in positions 2 and 4 on the Strat during quiet passages, there isn't much hiss, and practically no hum at all, as good as any cranked tube amp. When first turned on in the standby position, there is a slight hiss until the amp warms up and is switched on, then the hiss gets quieter. When the amp was new the fully cranked sound left a little to be desired, it had less usable range on the volume, but sounded good when dialed in just right.

  Next I changed the 7189 output tubes to Groove Tube EL84S Red Labels (made in the JJ factory), and rechecked the bias before trying them out. Again, with different tubes, I had to readjust the bias voltage to get back to 8.6 watts idle dissipation. That is why the adjustable bias is important, different tubes can have a very different idle wattage with the same grid bias voltage. The GTs were very similar in sound to the 7189s. The GTs were just a little warmer sounding, with slightly looser bottom-end, but still had very tight bass, were very bright, with clean mids and sparkling highs. With either brand of new tubes the amp sound is much more open, less boxy sounding. The amp now sounds much bigger and fuller. Of course a big part of the sound improvement is the new capacitors, as well as the lower bias and new tubes.

  Finally I got to A-B my amp with a friend's US made Blues Junior, and the difference was striking. The US made amps sound darker/warmer than the later MIM amps when new anyway, but close enough in sound for a fair comparison. With both amps' knobs all at 6 (except reverb at 1), my modified amp was slightly louder and much cleaner sounding. I could dial mine in to be just as warm and distorted as his unmodified amp, but the unmodified amp could not get as clean as mine without his being at a much lower volume. I am very happy with the results of the mods. I feel that I have the best of both worlds now, clean and driven sound, while still being able to get the same Blues Junior sound that I had to begin with, without as much of the boxy sound that it had. I also expect to get much longer output tube life with the idle wattage reduced from over 13 watts to 8.5 watts. I'd call it a win-win situation! I must give most of the credit to Bill Machrone for doing the homework (heavy lifting) and providing the quality parts. And thanks to Chuck for bringing over your amp and AB switch.

 You can see pics of the tubes here:  EL84 Tubes