Now I’m finally getting down to starting my restoration project I mentioned in my thread about getting a good multimeter.
I am an electronics noob, so I have no idea what to do about these 1960s sizes that seem to be rare these days.
Some Googling has given me a few thoughts:
Voltage can always be larger with no harm
Capacitance can always be larger for filter caps
If you stay within 20% of capacitance either way, you ought to be fine on other electrolytics
I’m cool with the voltage size, as that is a more of a limit.
What I’m not clear on is whether I really will be fine swapping in a 4.7uF cap for a 4uF cap, or a 150uF cap for a 130uf cap.
In searching the various parts sites, I see that they often have the specified values but they are either rare or expensive or both, so I definitely don’t want to pay $50 for a part that won’t ship until next summer.
You’ve already posted your answer: 4.7 is less than 4.8, (20%) and is bigger than 4.
Generally, a capacitor is specified as small as it can be, to fit into space and budget. And had to be small enough to get the voltage rating you wanted. So going bigger is not a problem.
The problems you can get by going bigger are probably not an issue with a Seeburg background music system anyway.
To clarify, I did the math and saw that 4.7 falls within 20%, but was looking for people who know this stuff to confirm that rule of thumb really is the case.
You’re right about the level of nuance here–we’re not talking about a Swiss watch. I have around 500 records for these machines, and they are all recorded in glorious 16⅔ rpm mono, so as long as I get a reasonable sound without terrible hum, I’ll be happy.
Thanks for the link. Their prices don’t seem crazy and it just might be easier to order from them then to wade through the listings of the regular parts suppliers.
Has hum been a problem? If so, you can make the power supply caps bigger, and maybe one of the early filter stages should go smaller. (I’m not going to look at it unless you tell me that hum is an actual problem).
Thanks! I haven’t even turned the electronic components on yet, but I will definitely take you up on that offer if hum is an issue.
My first attempt at desoldering was less than pleasant, as these were point-to-point, with 3-5 wires wrapped tightly around each post, so unsnarling that for all caps will take way too long.
I am probably going to do what some do and clip the leads close to the bad part, bend a hook in the end, and attach the new part to the old leads, one lead at a time, so I don’t get polarity messed up.
I did find a really cool PDF called “Seeburg music products semiconductor replacement guide” that is basically a list of alternatives for all of the transistors and other semiconductors. It’s clearly labeled “Replacements offered for Emergency Use Only” and the owner of the original scanned document had underlined that line in red pencil.
I also found another guide from a pinball machine repair website that gives all kinds of neat tips–the one I will use first is to put a 40W incandescent bulb inline with the line power cord in order to catch overcurrent situations before the damage occurs when I do the first startup.
This is a trade skill like playing tennis, or playing piano. I can’t do it either. I can only suggest that for me, buying more expensive de-soldering equipment makes it easier.
Are those all aluminum electrolytic caps? If so, I agree with the others that the capacitance value is (usually) not all that critical for them. If you can’t find the exact capacitance value, use a commonly-available cap that has a slightly larger capacitance. The same can be said for the voltage rating.
Well, not quite.
A long time ago I was working on a precision resistance bridge. While I was at it I decided to replace the main filter caps in the power supply. Not that they were bad. I just figured it was a prudent thing to do while I was in there, knowing that aluminum electrolytic caps tend to dry out after a couple decades.
I think there were four caps. I found some new caps that had the same geometry, pin spacing, and voltage rating, but the capacitance was around 50% higher. “No problem, more capacitance is a good thing!” I thought. After installing the caps and applying power to the unit, the hum in the 60 Hz transformer made a momentary “hum” sound that was significantly louder than before. I was also watching the filament in the glass bus fuse for the mains. I could see that it momentarily flexed quite a bit during power up.
Why did this happen? Because the larger caps increased the magnitude and duration of the inrush current. This put greater stress on the 60 Hz transformer, the mains fuse, and probably a bunch of other stuff.
At the very least I knew the fuse would probably fail soon. I ended up removing the (new) caps and replacing them with caps that had the same capacitance.
I checked my butt-load. I’ve got some 50’s and 100’s, but they are all 100v. A mixture of Marcap and Bennic. The ‘leads’ (is that what they are called?) are pretty short. They were cut out of discontinued cross-over assemblies. If you want them, PM me.
Got a bunch of 28’s and 31’s as well, if anybody wants them. I’ll NEVER uses these.
Good odds that those are non-polarized. Good for audio circuits, probably not so good for power supplies. Easy way to spot them is both ends are crimped if they’re axial leads. Radials will either be labeled NP or have no polarity marks.
And something similar can happen with the filter capacitors: A low frequency signal could drive the subsequent stage into distortion if the power supply doesn’t hold up, or even (in theory) overheat something by keeping the bias too high for too long.
But unless the system was so badly designed that it failed for that reason previously, 20% on the caps isn’t going to make it fail now – at least, not until after you’ve noticed it working so badly that you decide it’s unusable.
I had a look at the schematic. My goodness that is old school. From the days when transistors were costly precious devices.
A few thoughts.
The design is at least missing some of the evils of older circuits. It looks safe. No death caps or the like.
I’ll bet the unit is built with axial lead capacitors. This is going to be the biggest pain. Axials are rare beasts, as almost everything is made for PCB build and radial leads (or now SMD is making even leaded components fade away.) And you will find that modern capacitors are hilariously tiny compared to their older brethren.
As to values, as noted above, nearly everything is very loose tolerance. It looks more as if the unit was designed during the introduction of E series values, and uses a mix of decimal values and E6 values. Almost everything is available in E12 values, and using the nearest is fine. 4uF = 3.9uF. There is no sensible rule about not rounding down. The rule is to stay within tolerance band.
I don’t think there are any high tolerance requirements for capacitors. There are a few areas where there are 5% resistors which suggests someone was trying to make things tighter. But nothing that says be ultra careful with cap values. If there had been an RIAA equalisation stage you would have needed to be more careful.
Using non-polarised caps is going to be fine, especially for coupling capacitors. Curiously they are known to produce a lower distortion result. Not that this device will notice.
Desoldering. Invest in a big roll of solder wick. For big gobs of solder, a vacuum desoldering tool is useful as well. But solder wick will rip every last drop of solder out of a joint.
In late December I replaced the big fat power supply caps, and decided to see if that would be enough. I tried starting the machine with a light-bulb current limiter, and it worked fine without recapping everything else.
The mechanical part is up and running, but continues to give me grief–I’m winding my own springs at this point.
I really wanted to just play an audio feed through the amplifier so it would look cool but not require playing the records. This is actually supported by the device, so I got it working nicely in a few minutes. I kept it playing like this for some days.
Then… I noticed the bass started getting a little fuzzy, then more fuzzy. It had to be those old caps finally giving up their last.
I grumbled a bit and set the machine aside for several months. When I found the courage to open it up again, I but the bullet and replaced all of the caps I could, except for those 4uf ones, since they are all but impossible to access in the device.
I settled on a technique where I clipped the new cap leads very short and soldered 2" lengths of appropriate sized stranded wire to each one, with heat shrink tubing to cover the joints. This made it much easier to hook them up in the confines of the device.
Now it’s running perfectly! Thanks for all the help on this one.
