In general when repairing any amplifier in the tube era you should replace all of the paper caps as well as the electrolytic capacitors. In the thirty years of my experience on jukebox amps I have found that a bunch of the resistors are also out of tolerance. Just replacing the capacitors will most likely not bring you amplifier back to factory spec. When replacing components, use a solder sucker to remove the solder from the terminal, unwrap the defective component(s), and install the new parts. Heat the terminal and apply solder. Use just enough solder to coat the joint. More is not better. The finished joint should be smooth and bright. A dull solder joint is suspect as a cold solder joint which can lead to noise, cracks and pops and sometimes a loss of signal at that point in the amp. The bottom line is that a neat job will look good to your customer and it makes it easy to troubleshoot any other problems in the amp.

Testing tubes I don’t want to get into a discussion with the purist and "sonic quality" crowd, but, the facts are, a jukebox amp is not a piece of hi-fidelity gear. They have a lot of distortion and hum as well as having output transformers that are minimal at best. I use an industry classic standard tube tester that most other testers try to emulate. It is a military TV7-U. It wont lie to you and it is easy to calibrate. When I test a tube I test for shorts when it is just warming up as well as after it has warmed up for at least 5 minutes, then tap on the tube while checking for shorts when hot. I like to see the emission on the tube beat the minimum by a good margin. For dual tubes like a 12AX7 or 12AU7 I like to see the emission be within 20% of each other. The exception to this is for a phase splitter, I try to select a tube that is within 10% gain on either side or better. There are a few exceptions and they are 6CY7, 5879, rectifiers, 6L6 and other power tubes. If they are at the minimum or just a bit over they should be just fine.

Tube brands are another can of worms. If I can get a N.O.S. (new old stock) tube that’s great. I like the 12AX7 made in yugoslovia, the JAN 5814A/ 12AU7 made by phillips and the 6L6GB or 5881 made by sovtek. These tubes have excellent match and gain. The tubes I don’t much care for and won’t use are the sovtek 6V6, 6SL7, 6SN7, and 7199. They all suffer from one common problem, and, that is a whole bunch of heater to cathode leakage. Also, the pentode side of the 7199 is nowhere near like the real 7199. You will have to select a screen resistor to properly bias the triode or the audio will be very distorted as well as it may oscillate at about 50 Khz. If you put these into a jukebox amp you will think you have bad power supply caps. They are really awful. You can use the sovtek in a wurlitzer amp by making a negative bias network for the filaments of about 10 volts. It is possible to do this in a seeburg also but the bias network needs to be in the receiver. I hate modifying stuff like this and prefer to spring for the extra bucks to buy a N.O.S 7199.

MRA Series Amps are the easiest seeburg amps to rebuild. Most were made in Batavia NY by Sylvania. I am not sure who did the design, but, the Sylvania reps knew more about how these amps worked than any Seeburg engineer that I ever met. A problem started showing up with the amps that had avc (automatic volume) circuits in them. Seems that with the new bent needle in combination with the stereo and super cheap and thin promotional plastic records were causing the avc equipped amps volume to surge and "pump". Not too much of a problem in a raging honky tonk, but it scares the H*** out of people in an elevator as any californian can tell you that his record changer made such a noise during an earthquake. So the following mod was suggested by Sylvania, but, oddly not by seeburg.

MRA Series Mod # 1 off pin 3 of the avc rectifier change the 4.7 Meg to 3.3 Meg, change the 3.9 Meg to 2.7 Meg, remove C-20 &21 and replace them with a single 1.5 Mfd 250 Volt cap with the grounded end toward the 6J7. This solves the surging and pumping problem and makes the avc time constant longer.

MRA Series Mod # 2 Now that the amp could handle the stereo and promotional records well and high fidelity was the buzz word. It was thought that the bass cutoff of the seeburg amps were a bit too high. After all the new records could really put out the bass. Replace the 0.5 Mfd cap across the bass switch and the one from pin 3 of the 6SN7 with 0.56 Mfd caps. This lowers the bass corner a bit, but, not too much as to pick up turntable rumble.

MRA Series Mod # 3 "Hey man them o’l seeburg iron horse cartridges are a terrin up my new stereo records faster’n I can get-em-in, looks like somebody dusted them with corn starch, sound like S*** in bout a week". Enter the new and improved after market replacement cartridges. They were the Thorp, one made by Wico, and the best was the Pickering. All of these cartridges worked great mechanically, but, because they were true stereo cartridges wired internally as mono the output was only 30 Milli-volts. The original seeburg output was 50 Milli-volts. The Thorp and the Wico would balance, but the Pickering would not unless you ordered in the light weight counter balance used in the tormat series machines from Seeburg. A lot of operators just ordered in the whole tone arm assembly from Seeburg with a cartridge included. The other solution was to pull out the 6SK7 AVC tube to get more gain out of the amp. Will the mod is simplicity itself. Just remove the feedback resistor R-46 connected to pin 6 of the 6SL7 and replace it with a 180 Ohm resistor. Problem solved. This is all I ever do to all of the MRA series amps. These mods have served me and my customers well for years and years. If you do the above you will never get a complaint about lack of gain or bass.

Enter the SHFA series I am not sure who was the manufacturer of these amps. Could still be Sylvania, or RCA, not sure. This was the first step in modular design. These amps use small terminal boards mounted vertically. Really a good time saver for the manufacturing folks, but, for the service man they were a nightmare. They came in 2 flavors, code A and code B, also there was an occasional in between. What I started doing in the early 70’s was to remove the boards and replace all the parts with new ones. The only other thing that I do today is to use a matched set of (Voltage Forward) ECG 115 or 114 selenium diodes. If you use a silicon diode the avc won’t work right, nuff said. All of the transistors in these units were Beta matched. The color code on the transistor indicated what Beta range they were in. Just for grins we hooked up a sweep generator to the pre-amp board to check the RIAA equalization. To our surprise, it was way big time over compensated. Even more so with the code B boards. Further testing indicated that the higher the beta the wider the bandpass. Too much beta ,above 100 or so, and the circuit would oscillate at some ultrasonic frequency. Not a good design you say, well, not today, but back then you bought your dance card and necessarily you might not get to dance with the one who burng ya. You will find a lot of out of tolerance resistors in these amps. Take the time to replace them all. The bottom line is the amp really sounds great in the jukebox.

Speakers This is where I will get a little purist. Over the years I can’t tell you how many times we have told dealers and individuals to have the speakers re-coned. Not enough bass, scratchy sound, low volume, good sound on some days not on others, on and on. You can’t expect a paper cone speaker to last 30 years. change the caps out on those crossover networks also. The speakers in jukeboxes have a high SPL. A stiff paper cone and the efficiency goes to pot really fast, after all you only have a few watts to drive all of those speakers!.