68kMLA Classic Interface

This is a version of the 68kMLA forums for viewing on your favorite old mac. Visitors on modern platforms may prefer the main site.

Click here to select a new forum.
Anyone have an SE speaker?
Posted by: Phipli on 2024-03-17 02:34:14 Excellent 🙂 The bulk of the power ends up in R10. Good point, 3/4 of the power ends up in R10. What is R10s rating? Accounting for the power dissipation in R10 the speaker is within spec. header was 6.8V The supply to the transistors is between -12 and 12v - I'm surprised peak to peak was only 13.6V. Guess they have some overhead. I'm rusty on AC power calcs, it's a little while since I've done any so would need to dig out my databooks to check the numbers. Do I remember the frequency is pertinent though? What frequency was your sine wave?
Posted by: ymk on 2024-03-17 02:48:20 What is R10s rating? My best guess based on its size, 250mW. The supply to the transistors is between -12 and 12v - I'm surprised peak to peak was only 13.6V. Guess they have some overhead. No, the peak to peak was only 6.8V. I set the volume to 7 and played all kinds of sounds. 6.8V was the max Vpp the scope recorded. I'm rusty on AC power calcs, it's a little while since I've done any so would need to dig out my databooks to check the numbers. Do I remember the frequency is pertinent though? What frequency was your sine wave? I used a 100Hz sine, but it doesn't make a difference. I simulated the speaker as a resistive load, which is worst case. Real speakers' reactance increases around their resonant frequency and also increases with rising frequency:
Posted by: Phipli on 2024-03-17 02:56:14 No, the peak to peak was only 6.8V My bad - my phone wrapped the text so when I looked back up at what you wrote I saw "peak of" and missed the "peak to..." on th line before.
Posted by: maceffects on 2024-03-19 19:10:55 A youtuber says it's fine phil we should all sit down and know our places To be fair, the reason I shared the link is because JDW is an electrical engineer and I figured that might support the view that it’s perfectly acceptable and won’t hurt the sound circuitry of an SE. Of course, others had validated before I proposed the solution and I cannot see a material impact. If a speaker did fail, I’d certainly replace it.
Posted by: David Cook on 2024-04-13 16:33:08 I ran across this thread as I searched for a replacement speaker for my Macintosh 512K. It is a Regal 63 ohm 0.25W. Interestingly, my other 512K has a Samsung brand. I scrounged my junk box and found a 32 ohm speaker. It works fine, but is quieter (as expected). However, the speaker does not have the square mounting bracket. Hence, a cobbled-together set of hold-downs from washers.
Posted by: mattsoft on 2025-04-15 11:52:57 Ran across this thread looking for a replacement speaker for an SE/30 and unfortunately MacEffects is out of their speaker. So I ran across this model: Same Sky GA0576M. It looks similar to the MacEffects one and appears to physically fit according to dimensions. Shipping is more than the item, but such is life in 2025. Gonna order 1 and see if it works.
Posted by: David Cook on 2025-04-15 12:25:02 Ran across this thread looking for a replacement speaker for an SE/30 and unfortunately MacEffects is out of their speaker. So I ran across this model: Same Sky GA0576M. It looks similar to the MacEffects one and appears to physically fit according to dimensions. Shipping is more than the item, but such is life in 2025. Gonna order 1 and see if it works. Careful. It looks like that speaker is only 8 ohms.
Posted by: mattsoft on 2025-04-15 12:31:21 Careful. It looks like that speaker is only 8 ohms. yup, was gonna put a resistor inline to get it closer to 63 ohms. I just could not find a speaker with similar dimensions at 63 ohms. Heck, couldn't even really find anything at 63 ohms! lol!
Posted by: David Cook on 2025-04-15 12:46:24 couldn't even really find anything at 63 ohms Agreed! That's the problem I've been dealing with as well.
Posted by: adespoton on 2025-04-15 14:56:58 yup, was gonna put a resistor inline to get it closer to 63 ohms. I just could not find a speaker with similar dimensions at 63 ohms. Heck, couldn't even really find anything at 63 ohms! lol! Wouldn't that significantly drive down the dynamic range?
Posted by: mattsoft on 2025-04-16 09:04:43 Wouldn't that significantly drive down the dynamic range? dunno, but the OG speaker isn't exactly audiophile quality.
Posted by: s_pupp on 2025-04-17 05:56:41 If you are electronically inclined, you could consider an impedance matching audio transformer between the SE/30 logic board and the 8 ohm speaker. I myself do not have the knowledge to select the correct one. You could also try a simple LM386 audio IC powered by 5V, inserted between the SE/30 audio output and the speaker. There are pre-built amps with this IC available on eBay for under $3. It is such a simple circuit, that it could fit on a tiny board placed discretely somewhere near the speaker. It would likely need some EMF shielding - a tiny plastic box with copper foil on the outside of it might be enough. The LM386 is what I plan to do with my MacEffects 8 ohm speaker - I don't want to take any chances over-stressing my few remaining operational SE/30 boards. I have a couple of those IC's lying around that I purchased from my local Radio Shack in the early 1980's.
Posted by: ymk on 2025-04-17 08:44:59 There's no need to precisely match the 63 Ohms. Here's a list of 50-57mm speakers with 32-100 Ohm impedance. I'd start with this one at 57mm and 45 Ohms. 3D printing a mounting adapter isn't difficult.
Posted by: Callan on 2025-04-17 09:06:21 The LM386 is what I plan to do with my MacEffects 8 ohm speaker - I don't want to take any chances over-stressing my few remaining operational SE/30 boards. Yeah... I would't be super concerned about matching it exactly. I use a 16ohm speaker on the test bench to test any boards I work on. As for stressing chips... The Sony Chips don't drive the speaker directly (the part I'm assuming you're concerned about). The speaker is driven by U92 and 2A (push pull amp) with the TL071 at UA9 being the preamp. The TL071 is still readily available, and the transistors can be crossed super easy.
Posted by: shirsch on 2025-04-18 08:01:16 FWIW, I replaced a dead SE speaker with an 8-ohm unit and it's been working fine. One can do extensive circuit analysis or one can simply take a leap. I suspect the duty cycle is too short to cause problems. YMMV.
Posted by: tt on 2025-04-27 19:33:31 I personally wouldn't put a clear 8 ohm speaker in place of the original, aside from tech specs, the original looks better in a clear case and otherwise I don't want to alter the original timbre. I don't think I've seen anyone replace or refurbish the surround of these particular speakers, but it is a thing people do to restore vintage speakers (done it myself).
Posted by: shirsch on 2025-04-28 06:55:44 I've never heard of anyone doing a re-cone on tiny speakers, but stranger things have happened. FWIW, all the dead speakers I've encountered had open voice coil windings.
Posted by: David Cook on 2025-04-28 12:57:47 all the dead speakers I've encountered had open voice coil windings. Same here. What causes them to go open?
Posted by: shirsch on 2025-04-28 18:05:54 My theory is simple corrosion from exposure to humidity. There may be exposed copper where the winding leads come through the cone. It does seem particular to SEs as none of my older or newer compact Macs ever had speaker issues.
Posted by: LaPorta on 2025-04-28 19:02:44 My theory is simple corrosion from exposure to humidity. There may be exposed copper where the winding leads come through the cone. It does seem particular to SEs as none of my older or newer compact Macs ever had speaker issues. As a complete anecdote, all my machines have spent the vast majority of their lives in temperature and humidity controlled conditions. I’ve yet to have a speaker go bad.
< 2 >