Hi Nelson,
I go through life with no expectations these days. That way I'm always pleasantly surprised, or at least not disappointed. 🙂
I would appreciate it if you looked at Joe's questions and our answers to make sure Joe has been given correct information. Expanding on any answers would be a great chance to pick up some knowledge or another viewpoint.
Thanks Nelson.
You have the floor .......
-Chris
I go through life with no expectations these days. That way I'm always pleasantly surprised, or at least not disappointed. 🙂
I would appreciate it if you looked at Joe's questions and our answers to make sure Joe has been given correct information. Expanding on any answers would be a great chance to pick up some knowledge or another viewpoint.
Thanks Nelson.
You have the floor .......-Chris
My gosh I feel quite honored..
Okay.. Well; I didn't see your message and proceeded by checking the bias before I read it. So I guess I will report my findings.
With the unit dead cold I measure .56mv within about 2 seconds of power up on one side and the other side was .54.. I adjusted the .54 side to .56mv (I realize this was pretty insignificant but I wanted to see if I could do it anyway).. I SHOULD have replaced the pots as Jon from vintage recommended.. they are REALLY sensitive.. maybe this is "normal". It took me like 9 tries to get it and I did manage to spike it to 120mv for about 1sec and then I learned to power up really slow with the variac with very small adjustments.
Okay... so; now I need to follow the procedure posted by Mr. Pass and report back..
I will add that the "manual I have only has .33ohm emitter values in it for setting bias and mine are .20ohm.. the manual says if .33 emitters the cold bias should be .65mv so I suspect that it is close to being correct.
Okay.. back to the bench for more testing..and checking.
Okay.. Well; I didn't see your message and proceeded by checking the bias before I read it. So I guess I will report my findings.
With the unit dead cold I measure .56mv within about 2 seconds of power up on one side and the other side was .54.. I adjusted the .54 side to .56mv (I realize this was pretty insignificant but I wanted to see if I could do it anyway).. I SHOULD have replaced the pots as Jon from vintage recommended.. they are REALLY sensitive.. maybe this is "normal". It took me like 9 tries to get it and I did manage to spike it to 120mv for about 1sec and then I learned to power up really slow with the variac with very small adjustments.
Okay... so; now I need to follow the procedure posted by Mr. Pass and report back..
I will add that the "manual I have only has .33ohm emitter values in it for setting bias and mine are .20ohm.. the manual says if .33 emitters the cold bias should be .65mv so I suspect that it is close to being correct.
Okay.. back to the bench for more testing..and checking.
Okay... results:
At this point I suggest you do several things, the first being to
get some 1 and 2 amp fast blow fuses and put them in series
with the AC line.
DONE - WOULD RUN OFF 1 AMP BARELY AT 70V.. MUCH ABOVE THAT SHE'D BLOW.
Make up a pair of shorted male RCA connectors and put them
on the inputs so that the input has a grounded source but is
not otherwise attached to the outside world.
DONE INSTALLED
If they are opened (ohmmeter test) replace the resistors on the
output RC network with something comparable at about 5 ohms,
3 watts.
CHECKED BOTH AND THEY WERE BOTH OK @ 5.7ohms
It is not probable that you had bad RCA connectors. Usually if
even one of these has a connection, it will ground for both
channels. First, I would look for a bad internal input ground or
ungrounded cable shield. Check all your internal wiring twice
and try to confirm all connections with an ohmmeter. I don't
need to tell you to do this with the power cord unplugged.
CHECKED AS MUCH AS I COULD THINK OF.. I'LL KEEP LOOKING
Make sure that your output wiring has no proximity to the input
wiring.
DONE (1 was close on the left side)
After you have done all this, turn the amp on at low AC voltage
with no source or load, and note the voltage on the cases of
all the output transistors, both channels.
SET TO 70V
RIGHT SIDE - TOP TO BOTTOM
REAR
25.0V
25.1V
25.0V
11.3V
11.2V
11.2V
FRONT
24.7V
24.7V
24.7V
10.0V
10.7V
10.7V
LEFT SIDE - TOP TO BOTTOM
REAR
24.8V
24.7V
24.7V
10.6V
10.7V
10.6V
FRONT
24.7V
24.9V
24.9V
11.3V
11.2V
11.2V
I'll await your instructions.. THANKS!
Joe
At this point I suggest you do several things, the first being to
get some 1 and 2 amp fast blow fuses and put them in series
with the AC line.
DONE - WOULD RUN OFF 1 AMP BARELY AT 70V.. MUCH ABOVE THAT SHE'D BLOW.
Make up a pair of shorted male RCA connectors and put them
on the inputs so that the input has a grounded source but is
not otherwise attached to the outside world.
DONE INSTALLED
If they are opened (ohmmeter test) replace the resistors on the
output RC network with something comparable at about 5 ohms,
3 watts.
CHECKED BOTH AND THEY WERE BOTH OK @ 5.7ohms
It is not probable that you had bad RCA connectors. Usually if
even one of these has a connection, it will ground for both
channels. First, I would look for a bad internal input ground or
ungrounded cable shield. Check all your internal wiring twice
and try to confirm all connections with an ohmmeter. I don't
need to tell you to do this with the power cord unplugged.
CHECKED AS MUCH AS I COULD THINK OF.. I'LL KEEP LOOKING
Make sure that your output wiring has no proximity to the input
wiring.
DONE (1 was close on the left side)
After you have done all this, turn the amp on at low AC voltage
with no source or load, and note the voltage on the cases of
all the output transistors, both channels.
SET TO 70V
RIGHT SIDE - TOP TO BOTTOM
REAR
25.0V
25.1V
25.0V
11.3V
11.2V
11.2V
FRONT
24.7V
24.7V
24.7V
10.0V
10.7V
10.7V
LEFT SIDE - TOP TO BOTTOM
REAR
24.8V
24.7V
24.7V
10.6V
10.7V
10.6V
FRONT
24.7V
24.9V
24.9V
11.3V
11.2V
11.2V
I'll await your instructions.. THANKS!
Joe
Hi Joe,
Reading this would be easier if you used the "Quote" button on the right just over the message area. Right under where it says "COLOR". You can cut and paste into the copy field.
Like this ..........
Play with it and see what you think. 😉
-Chris
Reading this would be easier if you used the "Quote" button on the right just over the message area. Right under where it says "COLOR". You can cut and paste into the copy field.
Like this ..........
or, if you want to copy more, you have to go back and re-paste so that
the above has all the text, like this .......
Play with it and see what you think. 😉
-Chris
That's good news on the voltages. Have you confirmed the
voltages across each of the .22 ohm resistors?
.56 volts across the .22 ohm says 2.5 amps per device. Let's
take them all down to .1 or so while we figure this out. You
should then be able to operate the amp at 120 VAC with a 2A
fast blow fuse, if you power it up with a variac. If you just
plug the amp into the wall, the inrush will pop the fuse.
If you are concerned about the touchiness of the pot, you
can consider using a 10 turn type, but I suggest that we just
muddle through before making any actual improvements.
As I understand it, you have popped both RC output resistors,
implying that both channels have seen large high frequency
output. Is that the case, or this limited to one channel, and if
so is it the channel that had the proximity of input and output
wiring?
😎
voltages across each of the .22 ohm resistors?
.56 volts across the .22 ohm says 2.5 amps per device. Let's
take them all down to .1 or so while we figure this out. You
should then be able to operate the amp at 120 VAC with a 2A
fast blow fuse, if you power it up with a variac. If you just
plug the amp into the wall, the inrush will pop the fuse.
If you are concerned about the touchiness of the pot, you
can consider using a 10 turn type, but I suggest that we just
muddle through before making any actual improvements.
As I understand it, you have popped both RC output resistors,
implying that both channels have seen large high frequency
output. Is that the case, or this limited to one channel, and if
so is it the channel that had the proximity of input and output
wiring?
😎
Insight request
Hi Nelson,
Thanks for also jumping in to help out!
I have a basic request; would you be willing to add a bit of context as you request information from Joe?
It would be very interesting to follow your thought process while troubleshooting the amp and exactly what you're looking for at each step.
For example, why measure the collector voltages on each output device?
Also, wouldn't the voltages on the collectors as well as the emitter resistors change with the variac setting?
I think it would help myself and others understand the troubleshooting process a bit better and would be a bit of a tutorial for folks who have never done it before.
Thanks!
Hi Nelson,
Thanks for also jumping in to help out!
I have a basic request; would you be willing to add a bit of context as you request information from Joe?
It would be very interesting to follow your thought process while troubleshooting the amp and exactly what you're looking for at each step.
For example, why measure the collector voltages on each output device?
Also, wouldn't the voltages on the collectors as well as the emitter resistors change with the variac setting?
I think it would help myself and others understand the troubleshooting process a bit better and would be a bit of a tutorial for folks who have never done it before.
Thanks!
Nelson Pass said:
No I have not. Should I?
Okay I will.
Right side was popped left side was cooked but within specs. All lytics and coild have been replaced on both sides.. as well as all transistors (3 were bad on the left side). The proximity was not an issue until I rebuilt it and shortened the input cables (which I am kicking myself for doing now because it drastically reduced serviceability).. and it was not all that close.. the input was 1.2" away from the output at one point.. not parallel or anything.. I moved it away.. that was indeed on the left side however (the side that smoked the rc after the rebuild).
Everything SEEMED fine after all the rebuilding until I plugged in the source to 110v while the amp was powered.. it then buzzed and started to cook the left RC resistor.
Shall I disassemble and measure each .22?
I'll go ahead and set the bias to .1 and pause until I hear back.. Not sure what to look for once I do that.
Joe
Hi Joe,
I see you are enjoying the quote function! 😉
It does make the reading easier for the rest of us.
-Chris
I see you are enjoying the quote function! 😉
It does make the reading easier for the rest of us.
-Chris
Okay;
I made an assumption that you wanted to get the amp running on the 2 amp fuse; so I just simply tried bringing it up on the 2amp with a source connected.. it it comes up all the way and plays with the 2amp fuse sitting at 56mv cold bias.
I was monitoring DC on the right side with no speaker connected and noticed it swung wildly up to around 250mv while the left side sat at about 30mv. With a speaker connected it idles at 50mv cold and drops down to 30mv as it warms up.. the left side drops to about 20mv DC at the speaker output..
So; there definately is a difference between the channels in that regard. What bothers me is how it swings around with no load connected.. does it mean anything? Oh and this is with the inputs grounded as you suggested.
I played it for awhile @ 70v on the variac with nothing odd noted.
As you move the ac voltage up the DC voltage drops on the right side (not sure about the left). I then played it for about 20 min at 110v and noted a pretty consistant 30mv dc on the right channel..
Sorry; I know I am jumping ahead a little bit but I thought youd ask for this anyway.
I'm still waiting to hear if you'd like me to dissasemble and measure each emitter resistor.
I had the peak meters half way on some hungry JBL monitors and the 2 amp fuse did not blo.. a 1 amp did (I was curious ).
The heat sinks were evenly warming, but after 20 min they were barely warm.
j
P.S. I noted an error on my last post.. what I meant to say is all coil wound resistors were replaced on both sides.. as well as the lytics.
Here is a pic of one of the finished boards just before it went in
Front:
Back:
I made an assumption that you wanted to get the amp running on the 2 amp fuse; so I just simply tried bringing it up on the 2amp with a source connected.. it it comes up all the way and plays with the 2amp fuse sitting at 56mv cold bias.
I was monitoring DC on the right side with no speaker connected and noticed it swung wildly up to around 250mv while the left side sat at about 30mv. With a speaker connected it idles at 50mv cold and drops down to 30mv as it warms up.. the left side drops to about 20mv DC at the speaker output..
So; there definately is a difference between the channels in that regard. What bothers me is how it swings around with no load connected.. does it mean anything? Oh and this is with the inputs grounded as you suggested.
I played it for awhile @ 70v on the variac with nothing odd noted.
As you move the ac voltage up the DC voltage drops on the right side (not sure about the left). I then played it for about 20 min at 110v and noted a pretty consistant 30mv dc on the right channel..
Sorry; I know I am jumping ahead a little bit but I thought youd ask for this anyway.
I'm still waiting to hear if you'd like me to dissasemble and measure each emitter resistor.
I had the peak meters half way on some hungry JBL monitors and the 2 amp fuse did not blo.. a 1 amp did (I was curious ).
The heat sinks were evenly warming, but after 20 min they were barely warm.
j
P.S. I noted an error on my last post.. what I meant to say is all coil wound resistors were replaced on both sides.. as well as the lytics.
Here is a pic of one of the finished boards just before it went in
Front:
An externally hosted image should be here but it was not working when we last tested it.
Back:
An externally hosted image should be here but it was not working when we last tested it.
barchetta said:
I am a little concerned that we are clear on the voltages, as
in a previous post you mention .56 volts, and then 56 mV. As
I indicated, .56V is a lot of current, and I'm going to assume that
you meant 56 mV, which is fine.
Also, you need to make it clear which DC voltage you are
talking about - output DC or the DC across the emitter
resistors.
If the output DC is bouncing around a bit at 70 VAC, that is to be
expected, and if it doesn't wander beyond 50 mV at 110 VAC,
then things are looking up.
You did not mention whether both output RC network resistors
were fried in the last go-around. Was it one channel or both?
Given that the emitter resistors are not accessible during
operation, we can forego measuring them all, I think.
So at this point, the amp appears to work. I think you just might
want to run your JBLs for a few days with the Variac and 2A
fuse so as to catch out an intermittent problem.
After that, we will want to take the bias on both channels up a
little at a time until the amp still runs with the heat sinks at
45 deg C (just 10 seconds hot to the touch), increasing the
fuse size to 2.5A or 3A as necessary.
If you were lucky, then the problem may have simply been the
proximity of input and output wiring....
😎
Whats scary is that I dont know if it was MV or V.. That is very sad I know.. I will look closely at my multimeter tonight and report back. I am pretty sure it was mv though. When I refer to DC voltage I am talking speaker output.. so measured right before the speaker output fuse.. And when I measure with no speakers connected the right channel is all over the place.. my digital meter cant keep up.. I see it wonder all over between 20mv and 250mv.. (I hope its mv).. and it is rock solid on the left with no speaker.. but when I connect speakers it stays stead below 50mv.. as it warms it goes down to a steady 30mv at idle.. and when I play music it wanders just a little.. same on both sides but left side idles @ 20mv.. I think the above is tru @ 110vac input.. but I will re-verify tonight.
When I got the unit both rc's were cooked. When I did the little plug in cd player trick it was just the left side that smoked and that was the side where interconnects were in close proximity. Can you please explain why have the output and input wires in close proximity might cause this (or if someone else knows the answer chime in).. is it a 400a thing?
I will also start playing the unit with the 2amp fuse in tonight...
If it plays for a few hours with no issues where shall I set the bias across the .22ohm emitters to roughly?
And is it 10 seconds where you really dont want to hold on anymore cuz it will start to hurt a bit? And how thick is the skin on your hands? 🙂 Can you hold the tip of your solder iron for more than 500ms? 🙂
When I got the unit both rc's were cooked. When I did the little plug in cd player trick it was just the left side that smoked and that was the side where interconnects were in close proximity. Can you please explain why have the output and input wires in close proximity might cause this (or if someone else knows the answer chime in).. is it a 400a thing?
I will also start playing the unit with the 2amp fuse in tonight...
If it plays for a few hours with no issues where shall I set the bias across the .22ohm emitters to roughly?
And is it 10 seconds where you really dont want to hold on anymore cuz it will start to hurt a bit? And how thick is the skin on your hands? 🙂 Can you hold the tip of your solder iron for more than 500ms? 🙂
OK, I think we're on the same page.
No problem yet with the DC output voltages.
If one channel oscillates at full power at 100 KHz, there is a
reasonable chance that the other channel could pick that up
also, so let's not worry about that yet.
BTW, are you using any exotic loudspeaker cables (Litz wire,
coax cables, etc), and are your input cables the ordinary
shielded (radio shack) types?
When it comes time to up the bias, let's start with 50 mV and then
later work it up to 75 mV and from there we approach 100mv as
we aim for 45 deg C on the sinks after an hour.
The bias will drift up with temperature, so always set it at a lower
figure and wait an hour to see where it ends up. You have to do
this carefully, as a lot of blowups occur at this stage with just a
little slip. That's where the low value fuse comes in. I always
adjust to halfway between where I am and where I want to be
and then wait.
😎
No problem yet with the DC output voltages.
If one channel oscillates at full power at 100 KHz, there is a
reasonable chance that the other channel could pick that up
also, so let's not worry about that yet.
BTW, are you using any exotic loudspeaker cables (Litz wire,
coax cables, etc), and are your input cables the ordinary
shielded (radio shack) types?
When it comes time to up the bias, let's start with 50 mV and then
later work it up to 75 mV and from there we approach 100mv as
we aim for 45 deg C on the sinks after an hour.
The bias will drift up with temperature, so always set it at a lower
figure and wait an hour to see where it ends up. You have to do
this carefully, as a lot of blowups occur at this stage with just a
little slip. That's where the low value fuse comes in. I always
adjust to halfway between where I am and where I want to be
and then wait.
😎
Nelson Pass said:
In my test environment I am using ordinary zip cable.. Maybe monster brand I forget.. not shielded. For interconnects I am NOW using Acoustic Research cables.. I think their top line.. not sure.. bit the bottom as those are blue and mine are dark grey ($30 each at Best Buy).
When you say set the bias and mention going to 100mv are you talking hot or cold? The manual I have talks about cold settings only and using heat to determine the hot setting.. so can you clarify.. you rattled off a few numbers.. I assume you meant start at 50mv COLD.
I think I have a temp meter somewhere.. I'll start looking.
Any chance you could speculate why the right channel hunts around for voltage with no load (speakers connected).. its seems weird to me that the left does not do the same thing..
AND THANKS AGAIN!
OK on the cables.
When I give 100 mv for bias, I mean the value it settle into
after an hour or so.
If we had more equipment, we could proceed more rapidly, and
then again, maybe not considering my response time.
If you have a temp meter, fine, but we can nail this pretty
accurately with the skin test.
Humans have a consistent and abrupt response to temperature
on their skins. 40 deg C feels nice and toasty, at 45 deg you
can put your hand on it for a 10 second count, and at 50 deg
you are good for only a few seconds.
Correspondingly, I have consistently designed in the 40 to 50
deg C range, both because it is a good temperature for heat
sinks and also the skin test is a good metric when talking to
consumers and DIYers about heat sink temperature.
The issue with DC on the right channel is only important if the
channel is wandering when the AC line is up around 120VAC.
Somewhere below that the current sources have not stabilized,
and we do not have much expectation of DC accuracy.
😎
When I give 100 mv for bias, I mean the value it settle into
after an hour or so.
If we had more equipment, we could proceed more rapidly, and
then again, maybe not considering my response time.
If you have a temp meter, fine, but we can nail this pretty
accurately with the skin test.
Humans have a consistent and abrupt response to temperature
on their skins. 40 deg C feels nice and toasty, at 45 deg you
can put your hand on it for a 10 second count, and at 50 deg
you are good for only a few seconds.
Correspondingly, I have consistently designed in the 40 to 50
deg C range, both because it is a good temperature for heat
sinks and also the skin test is a good metric when talking to
consumers and DIYers about heat sink temperature.
The issue with DC on the right channel is only important if the
channel is wandering when the AC line is up around 120VAC.
Somewhere below that the current sources have not stabilized,
and we do not have much expectation of DC accuracy.
😎
Nelson Pass said:
Don't worry about the time.. if it takes you days to respond thats okay by me.. the other guys are impatient 🙂
IT WAS MV!! Thank god! I was in the 200mv range on my multimeter. I am going to order a scope off ebay if I can find one for the future.. sorry I dont have one now.
I will see how the DC looks at 120vac as you suggest and let you know if its wandering around. Otherwise I will work on the bias after it plays for a few hours set as is.
I'm starting to feel really optomistic that this thing is indeed fixed!
EDIT: I checked DC @ right speaker out and it does run from like 30mv up to 120mv and my meter cannot keep up.. this is with 120vac power.. BUT it only lasts for about 60 seconds.. and then it goes to a pretty steady 32mv. Is this alarming?
Uh boy... here is another one:
http://cgi.ebay.com/VINTAGE-THRESHO...6QQihZ008QQcategoryZ39783QQrdZ1QQcmdZViewItem
I hope that whomever buys it doesnt power it up though.. the LED display has come loose and is laying on the transformer! I can't help myself.. I might have to own a third!
Okay.. after 1 hour the bias voltage has dropped and the temp. is only 34.1c. I have no idea what increments I should move the bias voltage up.. I guess I will try 10mv at a time?
DC v at speaker outs is less then 10mv left and about 25mv right.. pretty damn good I think!
http://cgi.ebay.com/VINTAGE-THRESHO...6QQihZ008QQcategoryZ39783QQrdZ1QQcmdZViewItem
I hope that whomever buys it doesnt power it up though.. the LED display has come loose and is laying on the transformer! I can't help myself.. I might have to own a third!
Okay.. after 1 hour the bias voltage has dropped and the temp. is only 34.1c. I have no idea what increments I should move the bias voltage up.. I guess I will try 10mv at a time?
DC v at speaker outs is less then 10mv left and about 25mv right.. pretty damn good I think!
Well... last night was not fruitful.. I let it run for a few hours and it seemed fine... carefully moved the bias up to generate heat.. I found thermometer and taped the sensor to the heat sink.. it was running at about 34c while playing music quietly. I turned off the music and it dropped 2 degrees.. so I moved up the bias in increments that the pot would support (it is more like a voltage selector than a pot). I got it up to 40c and then I moved it up 15mv.. my hand felt like if it went up 5 more degrees just as you said I would only want it on there for 5-10 seconds.. so my hand was agreeing with the gauge. Well; there was a crackle and I noticed the woofer being extended... I quickly shut it down and set the bias back down to the previous level.. brought it up and after a few minutes... crackle.. and again woofer was being "oscillated".. so I measured DC at the speaker output.. and sure enough, it was doing the same thing I noted earlier when the speaker was disconnected.. sort of wildly oscilating .. I saw it hit 150MV but my meter could not really keep up..
I was very careful with the adjustment.. only adjsuting with power off because the pot is so touchy.. I'd adjust and then variac it up slowly until I made sure it did not exceed what I wanted and if it did I just shut it down and re-adjusted.. (can I blow some tube cleaner into that pot?).. I waited for the temp to even out before adjusting.. right or wrong I would use music to help it a little to its next stage.. Id turn off the music and note whether it would drop a little or if it would still rise.. soon as it dropped a little Id mute the source.
I do not know where I am cold now compared to the 56mv setting.. I went to bed at that point.. so I will measure what the cold bias is and also see if it takes a few hours to start this behavior again.. but I assume this is not good..
Definately not sure how to proceed at all. Help? 🙂
I was very careful with the adjustment.. only adjsuting with power off because the pot is so touchy.. I'd adjust and then variac it up slowly until I made sure it did not exceed what I wanted and if it did I just shut it down and re-adjusted.. (can I blow some tube cleaner into that pot?).. I waited for the temp to even out before adjusting.. right or wrong I would use music to help it a little to its next stage.. Id turn off the music and note whether it would drop a little or if it would still rise.. soon as it dropped a little Id mute the source.
I do not know where I am cold now compared to the 56mv setting.. I went to bed at that point.. so I will measure what the cold bias is and also see if it takes a few hours to start this behavior again.. but I assume this is not good..
Definately not sure how to proceed at all. Help? 🙂
Before departing for my "day job" I ran a quick cold test. Bias was @ 61mv and the oscillations if thats what they are called were there with the speaker connected.. crackles and the woofer would occasionally want to launch like the space shuttle. Damage has occured. 🙂
Not sure if I did it or time did it. I am guessing the DC fluxuations I saw with no load was indeed a warning sign? Anyway I see lots of erradic DC at the speaker output on the right side (same side I adjusted bias on)...
So I am wondering what might cause this? You mentioned that the I should watch for ocillations and pointed out that the transistors I chose were 2k not 4k.. I see on the schematic not it says clearly "4k only". Thats pretty clear 🙂 I didnt know what it meant until now. 🙂
Left side seems stable though... hmmmmmm
Not sure if I did it or time did it. I am guessing the DC fluxuations I saw with no load was indeed a warning sign? Anyway I see lots of erradic DC at the speaker output on the right side (same side I adjusted bias on)...
So I am wondering what might cause this? You mentioned that the I should watch for ocillations and pointed out that the transistors I chose were 2k not 4k.. I see on the schematic not it says clearly "4k only". Thats pretty clear 🙂 I didnt know what it meant until now. 🙂
Left side seems stable though... hmmmmmm