Amber Stereo 70

[mikeAtx]

Curious if anyone knows. My 70 has the silver front plate and black handles instead of the photo with black front plate and silver handles. Is mine a newer or older run?

 

[jbau]

IIRC, the black plates were the original, the silver came later. Back in the day, I always felt it was a dull/dark-sounding amp, punchy but dull.

 

[CalifonNJ]

In the spirit of "even a blind chicken finds a corn," replacing the LM391 on the failed side of the amplifier seems to have fixed it, at least after an hour or so of listening.

For someone like me with questionable skills, is there an easier way to measure the current at the speaker fuse (to adjust the bias, as recommended here) besides unsoldering the fuse?

I am still in DIY Audio Jail, so direct messages won't work until I can prove that I'm not a fraudster or spammer. I do respect the high bar set by DIY Audio but am not sure I can ever meet it.

 

[SpatialKing]

@CalifonNJ @indianajo
Sorry for the absence, apparently DIY wanted me to reset my account and I quit getting emails from there. I didn't know about that until today.

• About the fuse blowing - when a single channel blows in the Amber, it is 99.9% guaranteed that channel is blown. If the line fuse blows, then the 60W bulb trick is a good one and it may save the transformer from smoking. When you repair a channel, don't bother troubleshooting it, it is direct coupled and when it goes, it goes like dominoes. Replace the LM391 chip, the drivers and the output devices. There were two P/N's used for outputs and the MJ15003/15004 were the better of the two. DigiKey has them in stock.
• About leaky electrolytics on the power supply rail. It is unlikely a leaky electrolytic can pull the rail down. The smaller ones on the PCBA will explode first, the two big ones in the power supply will get hot and smoke. Most folks don't realize just how much overkill in the design of the power supply for the Stereo 70. Although the big electrolytics may blow, they will stink and smoke for a while before they go. Trust me, there will be no question in your mind when one of these guys goes! Buying caps with higher lifetimes at higher temps yields a better quality capacitor; likewise buying ones at higher voltages than needed. Avoid more than a 50% overvoltage though as the dielectric needs that voltage to fully saturate.
• About the 0.1 uFd caps, yes, I agree that sealed poly caps last a lifetime if not abused. The operative word here is sealed. The ones in the Amber are not sealed, so by now, they should be replaced. Use a polypropylene 0.1 uFd @ 100+ Volt cap as a replacement. Do find one that will fit on the board and replace all the 0.1 uFds.
• About the LM391-100 IC: Yes, National Semiconductor discontinued the part, not because it was a bad part, but because SS technology was moving into single chip amplifiers and those are less expensive than the LM391 semi discreet approach. A good argument here is whether the single chip approach is better than the original LM391 design, although the 391 does give more flexibility in the design and it was a very good design to begin with. The engineer who designed it was an audiophile, so he knew what he was doing. BTW, the LM391 has the ability to trade low frequency damping for a smoother midrange but I don't believe anyone took advantage of that feature and opted for a tighter bass. I had a long talk with the Marketing Director at National years ago at a CES show about extending the chip voltage range to 150V instead of 100V, but that would require a bigger chip in a 0.6 inch package and that package was too expensive (!) for manufacturing. So, a higher voltage part never came out. NTE made the LM391-90 part for a long time but I don't see it in there catalog now. I do have stock of LM391-100, as I followed up on the lifetime buy notice from National.
• Yes, in short, it is worth rebuilding a Stereo 70 with new polypropylene caps and setting the bias to ~ 35 mA. I do have the bias instructions in PDF format if anyone wants a copy. And forget the idea of more bias = better sound; the heat sinks are too small for that. It would take a heat sink three times the size and probably two outputs in series to handle the heat. That is a complete and total redesign.

 

[Junm]

Hi SpatialKing, i had done the modification you sent to me and it was good, After replacing two IC LM391-100 purchased from POLIDA and the mods, my amp was sound very good with a lot of Headroom. Thanks for the help.

 

[Alexandre]

• About the LM391-100 IC: Yes, National Semiconductor discontinued the part, not because it was a bad part, but because SS technology was moving into single chip amplifiers and those are less expensive than the LM391 semi discreet approach. A good argument here is whether the single chip approach is better than the original LM391 design, although the 391 does give more flexibility in the design and it was a very good design to begin with. The engineer who designed it was an audiophile, so he knew what he was doing. BTW, the LM391 has the ability to trade low frequency damping for a smoother midrange but I don't believe anyone took advantage of that feature and opted for a tighter bass.

Hi, I am about to try a few mods in an old amp (made in Brazil) that uses this chip.
About the trade-off you mention, I think you are referring to the "TRANSIENT INTERMODULATION DISTORTION" section of the datasheet. Open loop gain is reduced, but made flat out to 30Khz using 1MEG resistor between pins 3 and 9. Proposed by Matti Otala in his papers to mitigate TIM. Have you or anyone here compared it both ways? I guess I should try it myself but I also wanted to ask you fellows.

(BTW, others have stated that the only thing you need to do is have a decent amount of degeneration in the LTP, to mitigate TIM, and this chip seems to use 5K (!!!)... So, the high OLG with a low corner freq might be better...)

Also, has anyone used the chip in inverting mode?

Thanks,
Alex