0
\$\begingroup\$

Replacing electrolytics in a 30+ year old Ampeg BT-15 bass amplifier. The PS filter caps are \$ 2500 \mu F/80V\$ which are rare to be found; I know that 3300/100 would be okay for the filter. There's also a \$0.1 \mu F\$ in parallel which I asked about earlier and got helpful answers.

There is also a \$ 2500 \mu F/80V\$ electrolytic in parallel with a \$0.1 \mu F\$ just in front of the speaker connection, after the power transistor network. I know that it is in poor condition because pushing on it causes all kinds of static and popping. The replacement needs to be a through hole mount, preferably radial leads based on the configuration of PCB. I can't find that specific capacitor: the largest 80V cap is \$1800\mu F\$, and going to 100 V gives me either a \$2200\mu F\$ or \$3300\mu F\$ as options. Does it matter whether the replacement is higher or lower in capacitance? What effect on tone might either of these choichave?

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
1
  • \$\begingroup\$ Go to a speaker repair shop to buy bipolar capacitors for use on speaker lines. \$\endgroup\$
    – user105652
    Mar 15 '18 at 21:27
1
\$\begingroup\$

You don't show the schematics, but if the output is coupled via a capacitor, then I assume the amplifier is single-supply, which means the output voltage before the capacitor idles at half-supply relative to ground. Probably something like +30V or so, and it would swing between "close to 0V" and "close to 60V". Or maybe more volts, I'm assuming the designers used 80V caps for a 60V supply with a bit of a safety margin.

This means the cap doesn't need to be bipolar. If the output voltage of the amp is +30V at idle, then it isn't going to change sign, and the cap is always going to have a rather large DC voltage across it, and this voltage will always have the same polarity. So a standard polarized cap is fine.

(Please check the amp's output voltage before the cap: if it is is close to mid-supply then I'm right).

Since the output cap is the same as the cap used in the power supply, then it is safe to assume the manufacturer simply used the same value for convenience.

The value of the cap doesn't matter much* as long as it is large enough. A smaller cap will result in less bass, as its impedance increases at low frequencies. You can substitute with a larger cap. Your 3300µF 100V cap would be a good substitute for the 2500µF original cap. It will have a slightly lower impedance at low frequencies, so you'll get a bit more bass, something like 3dB.

If the loudspeaker enclosure was tuned for a specific value of cap, then using a different one may produce a small bump or a dip in the frequency response. It may sound a bit different. Caps aren't that expensive, so you can try and adjust to taste. This is a musical instrument, you should pick value that sounds best to you. That will probably be the largest.

\$\endgroup\$
2
  • \$\begingroup\$ Yes, it is single supply. This original cap is not bipolar. \$\endgroup\$
    – Bill N
    Jun 8 '18 at 0:47
  • \$\begingroup\$ Good! Then you can replace it with the 3300µF 100V. If the old cap was worn out and dry it would have lots of ESR, so replacing it with a new cap (with low ESR) will perhaps sound a bit different (but closer to the original sound the amp had when it was new). \$\endgroup\$
    – bobflux
    Jun 8 '18 at 12:20
1
\$\begingroup\$

In your diagram, the speaker is shown as a simple resistive load which is far from reality. In fact, the speaker is a reactive component (see Thiele-Small equivalent circuit). This implies that the whole output circuit has a (hopefully) well designed resonant characteristics, which in turn depend on the capacitance. I'd recommend choosing the capacitance as closely as possible. I believe that 2200uF should be well within the actual manufacturing tolerances (which are usually high for electrolytic caps).

The voltage rating should be not lower than 80V of course, but it also should not be much higher as well (like 250V or higher for example), but I don't think you could fit a high voltage cap in there anyway.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks. I've got plenty of room. The PS caps and this speaker cap are all 4 in long by 1.25 in diameter. But I think the 100 V cap will be okay, if not too short and thin for the existing mechanical supports. \$\endgroup\$
    – Bill N
    Jun 8 '18 at 0:46
0
\$\begingroup\$

Bigger V is better as well as C

Must be bipolar type

\$\endgroup\$
3
  • \$\begingroup\$ The original schematic shows an electrolytic. Please explain about why I should replace with a bipolar. I understand that the voltage is simply a maximum rating. Why is larger C better? \$\endgroup\$
    – Bill N
    Mar 16 '18 at 0:19
  • \$\begingroup\$ e-caps are normally polar and used in single supply PA's. For bipolar power, a bipolar cap must be used to decouple any DC offset in the output. A bigger cap lowers the breakpoint frequency but must be ultralow ESR to to improve damping factor. Direct Couple if possible is best. Big Bipolar caps are $$ \$\endgroup\$ Mar 16 '18 at 1:15
  • \$\begingroup\$ radioparts.com.au/product/06305980/… Get a bag of these, Put more in tandem and listen to the difference in Bass response. f(-3dB)=1/6.28RC for speaker R. I would also include 10uF bipolar and 0.1uF \$\endgroup\$ Mar 16 '18 at 1:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.