I guess 4th of July is being celebrated everywhere in the world, or my thread is not all that exciting...
Here are my further tests. After replacing Q202/204, and stepping back, and reconnecting JP201 ans 202, respectively and separately:
- reconnecting JP202 (R) turns off all alerts after the 4-5 seconds. This is through LBTester. Relays click and all that jazz.
- reconnecting JP201 (L) - which is where all of the extensive power transistor replacements occurred - leaves "Protection" on and no relays click. WTF is still wrong with this one?
This makes pretty literally no sense to me, given all work on this thus far.
Here are my further tests. After replacing Q202/204, and stepping back, and reconnecting JP201 ans 202, respectively and separately:
- reconnecting JP202 (R) turns off all alerts after the 4-5 seconds. This is through LBTester. Relays click and all that jazz.
- reconnecting JP201 (L) - which is where all of the extensive power transistor replacements occurred - leaves "Protection" on and no relays click. WTF is still wrong with this one?
This makes pretty literally no sense to me, given all work on this thus far.
As this thread is long and you have tried many things and --as you state on reconnecting the left power supply the protection stays "on " meaning its in operation due to a fault then could you check the current drawn in the "good " channel of both + and - power inputs ?
With this reference and using a variac with a current meter in series with each supply slowly increase the voltage and watch the current rising .
If it increases above the other good channel current used check whether its on the positive supply or the negative supply .
Or if no variac use a stablised power supply without connecting JP201
and slowly increase the voltage watching the current rise on the power supply or series current meter .
Just trying to narrow this down after all the advice and work you went through , no Rax I am not watching your celebrations.
With this reference and using a variac with a current meter in series with each supply slowly increase the voltage and watch the current rising .
If it increases above the other good channel current used check whether its on the positive supply or the negative supply .
Or if no variac use a stablised power supply without connecting JP201
and slowly increase the voltage watching the current rise on the power supply or series current meter .
Just trying to narrow this down after all the advice and work you went through , no Rax I am not watching your celebrations.
Rax,
I looked at the schematic for one second but it seems in this crazy design I see rails being regulated down over 30V meaning crazy power dissipation, meaning that in case of failure tube level high voltages could flood all your entire input circuitry. This failure has probably taken out a whole bunch of stuff. I would consider everything up for grabs at this point. When you rebuild this I would look at what parts to put in very carefully as well.
I would start by insuring the regulator circuits are in excellent working order on both channels.
I looked at the schematic for one second but it seems in this crazy design I see rails being regulated down over 30V meaning crazy power dissipation, meaning that in case of failure tube level high voltages could flood all your entire input circuitry. This failure has probably taken out a whole bunch of stuff. I would consider everything up for grabs at this point. When you rebuild this I would look at what parts to put in very carefully as well.
I would start by insuring the regulator circuits are in excellent working order on both channels.
Last edited:
I did more measurements last night - stayed up very late - and it actually seems at this point a lot of measurements are dead on (especially up to about half way through the circuitry, which is as far as I got last night). I am not sure why I get the different actuation of the "protection" circuitry, but given how well the voltages measure, it may be a very localized, hopefully last issue. After replacing the differential input pair on the suspect channel - which brought the voltages between channels to be very similar - as I assumed, my slightly elevated voltages (though similar) fell in place almost dead on after also reconnecting JP201/202. So this is really good news, definite progress on this saga.
Also, no smoke... Doing this through the LBT, but still, no further parts going fried. Given that B++ is at about 80V (due to drop over the LBT), 1/4W resistor could burn up easily on a shorted junction or whatnot.
Hoping to get some more bench time today and will "get back on there!" (Duncan?...)
Also, no smoke... Doing this through the LBT, but still, no further parts going fried. Given that B++ is at about 80V (due to drop over the LBT), 1/4W resistor could burn up easily on a shorted junction or whatnot.
Hoping to get some more bench time today and will "get back on there!" (Duncan?...
)grataku - thank you for your points.
A few responses:
- there's been a plethora of parts I changed (shorted Qs, burned resistors, etc.). There's been a burn out with this that took a lot of parts offline. I may still be chasing some of those, but I checked most transistors and replaced even some testing good on junctions but suspect for leakage or whatnot
- the rails are all good - especially the regulated ones, which had some issues
- It's also fully recapped at this point
Thank you, Duncan - it's a bit of a messier status here for the steps you suggest, I think. This is because I seem to experience some random actuation of the O/L alert, even without connecting the power stages. So I can't yet plug in the power stages (JP20x) and draw significant current to see, as even connecting just the JP402 or 403 can trigger O/L.
In a nutshell, I seem to see either 0V or about 17V in the collector of either Q219 and Q220 when their respective stages are powered (I only connect a limited number of JPs at a time).
Both have been changed (KSA992). I also changed Q213 and Q214 (KSC1845). L channel reacts better (9 out of 10 times no O/L; always P light goes off and relays engage). R not so much - O/L mostly on; relays may either not come on, or come on and then back off, in sync with P light.
What I've also been considering is disconnecting any protection circuitry and run the naked amp through the LBT. This would allow me to better see what triggers the protection circuitry - high output offset? Some other faulty condition, which I may be able to read through voltage-resistance analysis?
So I bit the bullet and ran it with all stages powered and got what I think is a reasonably dim bulb (75W, noticeably on, but yellow and dim).
Under these conditions, the Protection stays on and the relays don't engage. I have replacements for the relays, but they have worked occasionally (depending on other factors). But I feel at least I can run some measurements with the whole thing on.
For instance, the output offsets are very reasonable (under 20mV on one channel, under 1mV on the other).
To be noted, with the amp on through the LBT (again, 75W bulb) my B++ is 70V and the B+ is 45V. So there's some drop over the warm bulb. I'd be extremely interested if this is similarly so on a fully working amp and same bulb in the LBT.
Under these conditions, the Protection stays on and the relays don't engage. I have replacements for the relays, but they have worked occasionally (depending on other factors). But I feel at least I can run some measurements with the whole thing on.
For instance, the output offsets are very reasonable (under 20mV on one channel, under 1mV on the other).
To be noted, with the amp on through the LBT (again, 75W bulb) my B++ is 70V and the B+ is 45V. So there's some drop over the warm bulb. I'd be extremely interested if this is similarly so on a fully working amp and same bulb in the LBT.
Summary and a bit of regrouping
Alright, a bit of high level regrouping and summary of where I seem to be:
- not going into all details this far, but a whole array of parts have been replaced (including most power transistors on left side, O/L detect circuitry on both channels, input Qs on R, a bunch of zeners and a few other diodes, burned resistors, etc.)
- recapped almost fully
- modified the regulated rails
At this point, most voltages I measure are correct, except for an odd about 8V on the Q220/219 collectors. However, this doesn't trigger the O/L light, so I'm disregarding this for now.
I seem to get a dim orange glow from the LBT, so I think no dead shorts are present anymore. I believe the idle draw of this should be around 0.4A, so that sounds in the ballpark for the glow (I'd really want to get corrected if not so).
So, as I have the Protection light on, I am focusing on IC403. All voltage readings seem OK, except:
- pin 1: +.68V. The NAD SM calls for -.75V, though the TA7317P datasheet says +.75V. Anyone know what it should actually be? I get +.68V, so I may be good here.
- pin 6: a whopping +21.92V. This is the output, and I think it should be in the low Vs (+1.19V). Not sure what's going on here, but I'm currently vetting parts in that neighborhood and there's some things that are wrong. BTW - anyone can give me anything on the TS301, TS302 thermal relays? Haven't yet gotten much time to look into these.
Will post here as I find out more.
Alright, a bit of high level regrouping and summary of where I seem to be:
- not going into all details this far, but a whole array of parts have been replaced (including most power transistors on left side, O/L detect circuitry on both channels, input Qs on R, a bunch of zeners and a few other diodes, burned resistors, etc.)
- recapped almost fully
- modified the regulated rails
At this point, most voltages I measure are correct, except for an odd about 8V on the Q220/219 collectors. However, this doesn't trigger the O/L light, so I'm disregarding this for now.
I seem to get a dim orange glow from the LBT, so I think no dead shorts are present anymore. I believe the idle draw of this should be around 0.4A, so that sounds in the ballpark for the glow (I'd really want to get corrected if not so).
So, as I have the Protection light on, I am focusing on IC403. All voltage readings seem OK, except:
- pin 1: +.68V. The NAD SM calls for -.75V, though the TA7317P datasheet says +.75V. Anyone know what it should actually be? I get +.68V, so I may be good here.
- pin 6: a whopping +21.92V. This is the output, and I think it should be in the low Vs (+1.19V). Not sure what's going on here, but I'm currently vetting parts in that neighborhood and there's some things that are wrong. BTW - anyone can give me anything on the TS301, TS302 thermal relays? Haven't yet gotten much time to look into these.
Will post here as I find out more.
Protection circuitry checks
In a nutshell, we're back at this stage (never mitigated). I checked as many parts as I could from the protection circuitry (as I got tired to reuse the old transistors, I just replaced all those - Q401, 402, 403 - with KSC1845). I am getting the same readings, though. I will assume IC403 to be toast, so I am putting an order for some from Consolidated Electronics - reliable source in the US.
Welcome thoughts and feedback.
In a nutshell, we're back at this stage (never mitigated). I checked as many parts as I could from the protection circuitry (as I got tired to reuse the old transistors, I just replaced all those - Q401, 402, 403 - with KSC1845). I am getting the same readings, though. I will assume IC403 to be toast, so I am putting an order for some from Consolidated Electronics - reliable source in the US.
Welcome thoughts and feedback.
pwdiya12 - what sort of failures have you seen? Anything resembling my issues as depicted here?
I wonder if my orange glow of the LBT reflects this draw.
I enclose some further measurements from last night and this morning. I continue thinking the IC is toast.
I haven't gotten a lot of time to get back on the bench and work on this, other than discovering further previous tech misdeeds.
Being in series, it may have seemed OK to someone before me to reverse ZD301 and D309 in their positions, except that doing this obviously reverses the voltage over that pair...
Not sure what further damage this inflicted on the left channel. I assume this tech was addressing some other issue - maybe the catastrophic failure of the power transistors on that channel - and broke it a bit more in the process of fixing.
If anyone cares to pitch what the effect of this was on the amp - I'd be exceedingly grateful. I'm having a hard time regrouping and just seeing what the next step is here. I feel I should erase the blackboard and start anew, but it's really hard given the time spent with this amp.
Being in series, it may have seemed OK to someone before me to reverse ZD301 and D309 in their positions, except that doing this obviously reverses the voltage over that pair...
Not sure what further damage this inflicted on the left channel. I assume this tech was addressing some other issue - maybe the catastrophic failure of the power transistors on that channel - and broke it a bit more in the process of fixing.
If anyone cares to pitch what the effect of this was on the amp - I'd be exceedingly grateful. I'm having a hard time regrouping and just seeing what the next step is here. I feel I should erase the blackboard and start anew, but it's really hard given the time spent with this amp.