My SA-9500II is blowing fuses, dear God, please help....

[adam-ads]

Do these amplifiers have fuses on the speaker outputs? If not, would it be worth installing some?

 

[Big Bill]

I also enjoyed reading this thread and others like it.

I've learned so much from AK and the experts like Mark who give so freely of their knowledge helping keep our beloved vintage gear running.

Thanks Big Bill for starting this thread as I hope to tear into my SA 9500 II this winter and this has been great to follow along and read the out come.

Thanks Thomas!
Bill.

 

[Big Bill]

I tested incorrectly!

Mark, I tested incorrectly. I tested across the speaker instead of across the resistor.
I discovered this while contemplating your next set of instructions.

Here are test results of voltage tests across the resistors AND across the speakers:

Comparative voltage test of speaker VS. resistor readings. (Left and right channels)


Test CD. Across Speaker. Across Speaker. Across Resistor. Across Resistor.
Frequency Left Voltage Right Voltage Left Voltage. Right Voltage.
28 .3-1. .3-1 .1-.6 .2-.5
40 .4-1. .3-1. .1-.4 .1-.3
56 .4-1. .3-.8 .2-.6 .2-.7
79 .3-.6 .3-.7 .3-.9 .3-.9
112 .2-.6 .2-.7 .3-.8 .3-.7
159 .3-.6 .3-.7 .3-.6 .3-.7
220 .3-.6 .3-.6 .3-.5 .3-.6
320 .4-.6 .3-.6 .3-.7 .3-.6
450 .3-.5 .3-.5 .4-.6 .4-.6
630 .3-.4 .3-.5 .4-.6 .3-.6
890 .3-.4 .3-.4 .5-.6 .5-.6
1K26 .3-.4 .3-.4 .5-.6 .5-.6
1K78 .3-.4 .3-.4 .5-.6 .5-.6
2K5 .3-.4 .3-.4 .4-.5 .4-.5
3K6 .3-.4 .3-.4 .3-.4 .3-.4
5K .3 .3 .2-.4 .2-.4
7K1 .2 .2 .3 .3
10K .1 .1 .2 .2
14K2 .1 .1 .1 .1

Does this shed any new light on my situation?

Thanks in advance, Bill.

 

[Big Bill]

Do these amplifiers have fuses on the speaker outputs? If not, would it be worth installing some?

Wouldnt fusing them affect the signal quality?

Bill.

 

[markthefixer]

Bill, you followed my directions correctly BOTH times...

I revised the test.

By comparing the voltage drop across a "pure" resistance to the voltage drop across the speaker one can compute a ratio that will give you a good approximation of the speaker's impedance.

When the voltages are equal, the resistance (8 ohms) and speaker impedance are equal.

It will take a while for me to crunch the numbers, I'll probably suck them into a spreadsheet sometime soon when I'm feeling bored...

BTW, the meter seems to poop out above 3 khz or so.... no biggie...

 

[Big Bill]

New cabling and no resistor...

WE have one unassailable and irreconcilable fact here - blown amp channels.... Have you inspected the entire length of the spliced cable minutely - for insulation damage or evidence of chewing? And compared them side by side?

I have inspected the cables for damage and have found none. Nevertheless I
have "replaced" them with new 16 Gauge cable (all I can afford at the moment).

I have retested the amp voltage (across speaker) with the new cabling but WITHOUT the inline resistor. Here are those readings for the SA-9500II @ -34 DB volume setting using the same CD test tones as before:

Freq left right
28 .5-1. .4-1.
40 .5-1. .4-1.
56 .5-1. .6-1.
79 .6-1. .6-1.
112 .5-1. .7-1.
159 .7-1. .8-1.
220 .6-1. .6-1.
320 .7-1. .7-1.
450 .7-1. .7-1.
630 .8-1. .8-1.
890 .8-1. .8-1.
1k26 .8-1. .8-1.
1k78 .8-1. .8-1.
2k5 .8-1. .7-.8
3k6 .7-.8 .7-.8
5k .6-.7 .6-.7
7k1 .5 .5
10k .3-.4 .3-.4
14k .2 .2

What do you think about this?

Gratefully, Bill.

 

[markthefixer]

I have inspected the cables for damage and have found none. Nevertheless I
have "replaced" them with new 16 Gauge cable (all I can afford at the moment).

I have retested the amp voltage (across speaker) with the new cabling but WITHOUT the inline resistor. Here are those readings for the SA-9500II @ -34 DB volume setting using the same CD test tones as before:

Freq left right
28 .5-1. .4-1.
40 .5-1. .4-1.
56 .5-1. .6-1.
79 .6-1. .6-1.
112 .5-1. .7-1.
159 .7-1. .8-1.
220 .6-1. .6-1.
320 .7-1. .7-1.
450 .7-1. .7-1.
630 .8-1. .8-1.
890 .8-1. .8-1.
1k26 .8-1. .8-1.
1k78 .8-1. .8-1.
2k5 .8-1. .7-.8
3k6 .7-.8 .7-.8
5k .6-.7 .6-.7
7k1 .5 .5
10k .3-.4 .3-.4
14k .2 .2

What do you think about this?

Gratefully, Bill.

Without the resistor's "reference" resistance and the voltage across the resistor at the same time as reading the speaker voltages, the numbers are meaningless... just the voltage across the speakers, with NO idea of the current draw.... basically these are just VU meter readings of what the AMP is sending to the speakers, not what the speakers are demanding from the amp.

As for the rest of it, long term I would find a mass to sit the TT on, because you don't want a dead amp just because the subsonic filter got turned off. Lots of DIY solutions that are inexpensive and can have WAF built in.

 

[Big Bill]

Thanks for your reply mark.

I have three questions I’m hoping you will answer.

Firstly, what is the least amount of sustained DC resistance the SA-9500II will safely tolerate from the speakers?

Secondly, what are the model numbers of the two types of replacement transistors I need to repair the “donor” amp?

Thirdly, which supplier should I purchase these transistors from?

Thanks once again for all your help Mark!

 

[markthefixer]

Thanks for your reply mark.

I have three questions I’m hoping you will answer.

Firstly, what is the least amount of sustained DC resistance the SA-9500II will safely tolerate from the speakers?

Secondly, what are the model numbers of the two types of replacement transistors I need to repair the “donor” amp?

Thirdly, which supplier should I purchase these transistors from?

Thanks once again for all your help Mark!

That's a whole nuther story...

edit: you might want to read this thread, it's only one page long and has pictures of what we may do to fix your amp....
http://www.audiokarma.org/forums/showthread.php?p=1066827

You also want to read the thread I link to in the other post.

I am considering using the Fairchild FJP5200/FJP1943 as outputs on the 9500II. Fairchild is making a series of copies of the Toshiba 2SC5200/2SA1943 transistors in TO-264, TO-3P, and TO-220 packages. The FJP pairs are the TO-220's. They are 250V, 15A, 80W, 30MHz transistors. Digikey carries both the R (hFE 55-110) and the O (hFE 80-160) grades for about $1.00 each.

Thank You for the information, it is greatly appreciated - somehow I missed these while searching through Fairchild (and mouser) again the other day to see if I could find something new that was useful. And I also find them on MOUSER as well, and I was using THEIR search engine for high powered TO-220 transistors!! grrrrr........
The 30mhz might take some taming, but if the originals were 14mhz, it shouldn't be much of an issue.
But it does make it a hands on project - to have scopes etc on hand to check for nasty consequences.
Going down in frequency doesn't usually cause problems - with the exception of the Pioneer NSA amplifier's driver and output transistors - THAT results in a distortion penalty.

(more powerful to-220 from batwing replacements)
512-FJP1943OTU pnp to-220 bce 230v 15a 80w 30mhz 80-160hfe $1.21
512-FJP5200OTU npn to-220 bce 230v 15a 80w 30mhz 80-160hfe $0.85


You will have to make (or get) the hold downs with the solder lug collector connections, as well as longer metric screws to compensate for the height difference - I don't remember if I used 20, 25 or 30mm long screws.

And then there is the need to check for problems using faster transistors....

 

[Big Bill]

Thanks Mark!

And then there is the need to check for problems using faster transistors....

Oh boy, how is that done? :sigh:

P.S. I'm still wondering how few Ohms speaker resistance the SA-9500II will tolerate long term, do you have any idea what it is?

 

[markthefixer]

Thanks Mark!



Oh boy, how is that done? :sigh:

P.S. I'm still wondering how few Ohms speaker resistance the SA-9500II will tolerate long term, do you have any idea what it is?

There IS no easy answer for the ohms question. The less ohms, the greater the current demand, and for an amp that can drive 8 ohms, driving 4 ohms or less the heat generated by that current demand is what kills the amp. It either becomes more than the TRANSISTOR can dissipate,
OR
more than the HEAT SINK can dissipate.

The transistor speed increase checking is wizard territory.

original sa-8500ii
2sb617a pnp batwing bce 120v 6a 70w 80hfe 14mhz
2sd587a npn batwing bce 120v 6a 70w 80hfe 14mhz

original sa-9500ii
2sb618a pnp batwing bce 150v 7a 80w 80hfe 14mhz
2sd588a npn batwing bce 150v 7a 80w 80hfe 14mhz

But I used the:
863-MJE15033G pnp to-220 bce 250v 8a 50w 30mhz hfe:70 $0.87
863-MJE15032G npn to-220 bce 250v 8a 50w 30mhz hfe:70 $0.87

which is an increase from 14 to 30mhz without problems, so we are considering using:
512-FJP1943OTU pnp to-220 bce 230v 15a 80w 30mhz 80-160hfe $1.21
512-FJP5200OTU npn to-220 bce 230v 15a 80w 30mhz 80-160hfe $0.85

so we may not get in trouble. but without a scope etc it's hard to determine if an amp is breaking into RF oscillation, until it goes up in smoke..

 

[SteveM]

:yes:

Hello guys,
I have a question that I need your opinion on. . I was playing some music on the Empire TT I’ve been using recently and as I walked by the speakers I observed the woofer pumping back and forth violently. I went to the amp and the right side felt much warmer than the left side. I posted a question to the TT forum and learned it was TT “rumble”. I also learned what the 15 HZ filter on my amp is for. Afterward I flipped the 15 HZ switch, played a few albums and “tested” to see if the amp was hotter on the right. I found the heat to be even on both sides of the amp. Before I got this “solid base” Empire I had been using my B&O TT’s, both of which I learned have a spring type suspension which might be why I hadn’t noticed this “rumble” before.

My question is since the Empire is "new" and it was the source with both amps, is it possible “rumble” caused some kind of feedback loop and caused the amps to blow a channel.

Bill.

Entirely likely, Bill. You also stated somewhere that this problem
showed up after moving the physical location of the speakers.

If the new speaker location relative to the turntable places the TT
in a 'node' where some frequency displays a resonance maximum
(positioning speakers in a corner, especially in a room prone to
hard reflections, like a basement, is a good formula for this),
AND the turntable has a solid base, which does not isolate it
from low-f vibrations, AND the LF filter was switched out...

well, to repeat - entirely likely. The 'pumping' action you saw is
also indicative of this. And high-amplitude physical excursions of
the speaker cones point to higher low-frequency signal output
from the amp which could indeed have stressed the output stage to failure.
High listening volumes could also contribute, on top of this.
The fact that only one channel of the amp was getting warmer
(think - was it the corresponding channel's speaker that was
'pumping'?) points to that speaker's postion being the one
'feeding' the feedback. I could provide you with some data
regarding speaker placement in a room, based on room dimensions,
that was intended to be used to optimize base response by either
optimizing or reducing low-frequency node amplitudes in the room
(known as 'tuning the room').

We tend to think of 'feedback' as producing that ever-rising 'wail'
so common in rock performances (and used so artistically in some
cases - Hendrix comes to mind), but low-frequency acoustic feedback
DOES exists - those Rumble Filters were designed in specifically to
ameliorate this.

So...yes, it looks like you may have figured out your own
problem...good going, Mr. Holmes.

I know how unnerving it can be 'not knowing' what caused
a failure...it's hard to just relax and enjoy listening to the
music after that...

 

[SteveM]

More SA-9500 ii stuff -

Thought I'd add a little info...

I ended up here because I picked up one of these amps
at a thrift shop cheap, and bought it based entirely on
weight. I was extremely pleasantly surprised to find that
it has essentially 'dual-mono' construction inside:
separate big, beefy transformers for each channel,
power supply filter caps, etc.

The only thing that I was not impressed by was the
(in my opinion) undersized heatsinks. No worries -
I'm planning a complete rebuild (cap upgrade, output
stage upgrade, etc.) and the present location of
all relevant circuitry lends itself well to place nice,
huge extrusions on the sides of the amp, with nice,
big aluminum flanges leading back inside. Probably
add a hand-finished wooden case, while I'm at it.

Many thanks to Mark for posting on AK the source for
a service manual at http://www.kallhovde.com
- that will no doubt make this project go much
more smoothly. I will keep you guys up to date
with nude pictures and running commentary.

Again, kudos to Mark, and congrats to Big Bill.:thmbsp:

 

[Big Bill]

There IS no easy answer for the ohms question. The less ohms, the greater the current demand, and for an amp that can drive 8 ohms, driving 4 ohms or less the heat generated by that current demand is what kills the amp. It either becomes more than the TRANSISTOR can dissipate,
OR
more than the HEAT SINK can dissipate.

This, I can understand.

The transistor speed increase checking is wizard territory.

original sa-8500ii
2sb617a pnp batwing bce 120v 6a 70w 80hfe 14mhz
2sd587a npn batwing bce 120v 6a 70w 80hfe 14mhz

original sa-9500ii
2sb618a pnp batwing bce 150v 7a 80w 80hfe 14mhz
2sd588a npn batwing bce 150v 7a 80w 80hfe 14mhz

But I used the:
863-MJE15033G pnp to-220 bce 250v 8a 50w 30mhz hfe:70 $0.87
863-MJE15032G npn to-220 bce 250v 8a 50w 30mhz hfe:70 $0.87

which is an increase from 14 to 30mhz without problems, so we are considering using:
512-FJP1943OTU pnp to-220 bce 230v 15a 80w 30mhz 80-160hfe $1.21
512-FJP5200OTU npn to-220 bce 230v 15a 80w 30mhz 80-160hfe $0.85

so we may not get in trouble. but without a scope etc it's hard to determine if an amp is breaking into RF oscillation, until it goes up in smoke..

This is confusing!

 

[Big Bill]

:yes:


Entirely likely, Bill. You also stated somewhere that this problem
showed up after moving the physical location of the speakers....

Yes! I moved the speakers to another wall in the room, which put the tt directly across the room from the left speaker.

If the new speaker location relative to the turntable places the TT
in a 'node' where some frequency displays a resonance maximum
(positioning speakers in a corner, especially in a room prone to
hard reflections, like a basement, is a good formula for this),
AND the turntable has a solid base, which does not isolate it
from low-f vibrations, AND the LF filter was switched out...

I believe it did.

well, to repeat - entirely likely. The 'pumping' action you saw is
also indicative of this. And high-amplitude physical excursions of
the speaker cones point to higher low-frequency signal output
from the amp which could indeed have stressed the output stage to failure.
High listening volumes could also contribute, on top of this.
The fact that only one channel of the amp was getting warmer
(think - was it the corresponding channel's speaker that was
'pumping'?) points to that speaker's postion being the one
'feeding' the feedback.

I believe your right!

I could provide you with some data
regarding speaker placement in a room, based on room dimensions,
that was intended to be used to optimize base response by either
optimizing or reducing low-frequency node amplitudes in the room
(known as 'tuning the room').

Please do!

We tend to think of 'feedback' as producing that ever-rising 'wail'
so common in rock performances (and used so artistically in some
cases - Hendrix comes to mind), but low-frequency acoustic feedback
DOES exists - those Rumble Filters were designed in specifically to
ameliorate this....

I see now!

So...yes, it looks like you may have figured out your own
problem...good going, Mr. Holmes..

Thank you!

I know how unnerving it can be 'not knowing' what caused
a failure...it's hard to just relax and enjoy listening to the
music after that...

Amen, you sound like youve been there before!

Gratefuly, Bill.

 

[Big Bill]

Thought I'd add a little info...

I ended up here because I picked up one of these amps
at a thrift shop cheap, and bought it based entirely on
weight. I was extremely pleasantly surprised to find that
it has essentially 'dual-mono' construction inside:
separate big, beefy transformers for each channel,
power supply filter caps, etc.

The only thing that I was not impressed by was the
(in my opinion) undersized heatsinks. No worries -
I'm planning a complete rebuild (cap upgrade, output
stage upgrade, etc.) and the present location of
all relevant circuitry lends itself well to place nice,
huge extrusions on the sides of the amp, with nice,
big aluminum flanges leading back inside. Probably
add a hand-finished wooden case, while I'm at it.

Many thanks to Mark for posting on AK the source for
a service manual at http://www.kallhovde.com
- that will no doubt make this project go much
more smoothly. I will keep you guys up to date
with nude pictures and running commentary.

Again, kudos to Mark, and congrats to Big Bill.:thmbsp:

Good luck Steve! I for one, anxiously await posts of your progress!

Again a special thanks to markthefixer for all his help! Thanks also to the rest of you who have posted on this thread!

Greatfuly, Bill.

 

[sipuser]

Check out post #25 which describes this issue of over clocking.

http://www.audiokarma.org/forums/showthread.php?t=260131

 

[SteveM]

SA-9500 Output transistors

Thanks to EchoWars (Warren Zevon rules!) for his info on ST Microelectronics'
bipolar output transistor data, and to sipuser for posting it here.

I can appreciate Big Bill's trepidation about not using exact replacement parts.

The SA-9500 that I picked up at a local thrift is not powering up AT ALL,
which usually points to the internal protection fuses being blown, most
often by a shorted output stage. I'll no doubt be having to use subs for the
output transistors, and will keep you guys posted on what works out.

Long story short, when I get through finishing up on the Hitachi HA-1100
I'm working on (love these vintage 70's beasties!), I'll start on the 9500,
and post progress here.

 

[Bob91343]

Where can I buy 2SB618A transistors, or substitute?