First off, many thanks to those who helped my find my grounding issue. Once the PSU board was isolated from the chassis by switching all of the standoffs to nylon, and I used a shorting plug when resetting the bias and offset, all was working well.
But every time I think I have finished building the amp, it seems I find a new problem... I hope it is appropriate for me to post this here; if it warrants its own thread please let me know and I'll be happy to comply.
I have been using an AVM preamp with balanced out going to the J, and since resolving my previous issue, everything has worked fine. Needing a second balanced input, which the AVM is lacking, I bought a new Schiit Freya N. After connecting the Freya in place of the AVM, I powered it up and when it came out of standby, fired up the Aleph J.
Immediately a loud hum came from the speakers (volume knob still at minimum), which suddenly stopped and was followed by a burning smell. The 3A fuse in the J had blown, but there was no other visible damage. After replacing the fuse I checked the bias and offset and both were fine.
[ It occurred to me later that the Freya was in passive mode (the default out of the box) when the fuse blew so there was no gain or active circuit involved - which makes this all the more puzzling to me. ]
I hooked the AVM back up, and all was ok again. Try with Freya back in (with my finger in the J's power switch) and the loud hum was back. So clearly, there is an issue with the Freya, right?
I remembered that I had recently acquired a retired Crest Audio pro audio amp, and hooked it up to the Freya's balanced output - and it played music just fine.
So, I'm at a loss as to where to start looking. I suspect a DC offset on the Freya's balanced out, but don't know how to confirm that. I was about to request an RMA but having confirmed that the Freya works with a different amp, I'm having second thoughts about that. Any pointers are welcome and appreciated.
But every time I think I have finished building the amp, it seems I find a new problem... I hope it is appropriate for me to post this here; if it warrants its own thread please let me know and I'll be happy to comply.
I have been using an AVM preamp with balanced out going to the J, and since resolving my previous issue, everything has worked fine. Needing a second balanced input, which the AVM is lacking, I bought a new Schiit Freya N. After connecting the Freya in place of the AVM, I powered it up and when it came out of standby, fired up the Aleph J.
Immediately a loud hum came from the speakers (volume knob still at minimum), which suddenly stopped and was followed by a burning smell. The 3A fuse in the J had blown, but there was no other visible damage. After replacing the fuse I checked the bias and offset and both were fine.
[ It occurred to me later that the Freya was in passive mode (the default out of the box) when the fuse blew so there was no gain or active circuit involved - which makes this all the more puzzling to me. ]
I hooked the AVM back up, and all was ok again. Try with Freya back in (with my finger in the J's power switch) and the loud hum was back. So clearly, there is an issue with the Freya, right?
I remembered that I had recently acquired a retired Crest Audio pro audio amp, and hooked it up to the Freya's balanced output - and it played music just fine.
So, I'm at a loss as to where to start looking. I suspect a DC offset on the Freya's balanced out, but don't know how to confirm that. I was about to request an RMA but having confirmed that the Freya works with a different amp, I'm having second thoughts about that. Any pointers are welcome and appreciated.
And the FIFA goal of the year was scored by a striker with only one leg, period!
@beeah
Make sure the XLR wiring connection is the same at Aleph J end, and at the preamp end. I'd look at how the hot, cold and shield are connected at Aleph J end, and then I'd check to make sure it's exactly the same at Freya's end.
Any DC that may be present at Aleph J input, could be measured from its XLR connectors - just connect and power up Freya, DO NOT power up Alep J... and measure/look for any DC at Alep J XLR pins (remove the top cover).
Have a look at my latest posts here on this thread (last couple of pages) ... they may be useful.
Also, I see that you are using a third-party PS.... which we do not know much about.
Lastly, you have to give us detailed, well-focused and well-lit photos that show everything... especially about that PS PCB. Its schematics wouldn't hurt either... if you have it. Or, the photos of the unpopulated PS PCB - both sides.
Make sure the XLR wiring connection is the same at Aleph J end, and at the preamp end. I'd look at how the hot, cold and shield are connected at Aleph J end, and then I'd check to make sure it's exactly the same at Freya's end.
Any DC that may be present at Aleph J input, could be measured from its XLR connectors - just connect and power up Freya, DO NOT power up Alep J... and measure/look for any DC at Alep J XLR pins (remove the top cover).
Have a look at my latest posts here on this thread (last couple of pages) ... they may be useful.
Also, I see that you are using a third-party PS.... which we do not know much about.
Lastly, you have to give us detailed, well-focused and well-lit photos that show everything... especially about that PS PCB. Its schematics wouldn't hurt either... if you have it. Or, the photos of the unpopulated PS PCB - both sides.
Pretty standard CRC Aleph PS. This is the actual board: BrianGT Mini-A PSU
0.47 ohm Rs in parallel; bleeders; 10000uF/25V caps; OnSemi MUR2020 diodes. I did not populate the LED components.
I will post pictures later, as well as voltage measurements on the inputs.
0.47 ohm Rs in parallel; bleeders; 10000uF/25V caps; OnSemi MUR2020 diodes. I did not populate the LED components.
I will post pictures later, as well as voltage measurements on the inputs.
@Extreme_Boky
No pictures yet, but I verified connection of wiring between Aleph XLR jack and both channel boards (Pin 1 to signal ground; Pin 2 to In+; Pin 3 to In-) and everything is correct, with no shorts between pins. If this is not what you meant in your first paragraph, please clarify.
No DC between input pins on either channel with Freya powered up.
Regarding your recent post about grounding, on these boards common ground is tied to both output and input grounds. I have board common ground wired to PSU ground. I believe this is equivalent to the wiring scheme you suggested. Also, with nylon standoffs there is no longer continuity between PSU board and chassis.
I will post pictures (didn’t have sufficient light last night), but my next troubleshooting steps are:
1) test with one channel connected at a time to determine if the problem is channel or system related
2) disconnect secondary output devices on both channels (what this means will be clear once I post pictures) to possibly determine whether bad MOSFET(s) is to blame.
No pictures yet, but I verified connection of wiring between Aleph XLR jack and both channel boards (Pin 1 to signal ground; Pin 2 to In+; Pin 3 to In-) and everything is correct, with no shorts between pins. If this is not what you meant in your first paragraph, please clarify.
No DC between input pins on either channel with Freya powered up.
Regarding your recent post about grounding, on these boards common ground is tied to both output and input grounds. I have board common ground wired to PSU ground. I believe this is equivalent to the wiring scheme you suggested. Also, with nylon standoffs there is no longer continuity between PSU board and chassis.
I will post pictures (didn’t have sufficient light last night), but my next troubleshooting steps are:
1) test with one channel connected at a time to determine if the problem is channel or system related
2) disconnect secondary output devices on both channels (what this means will be clear once I post pictures) to possibly determine whether bad MOSFET(s) is to blame.
I wouldn't touch MOSFETs at this stage.... it seems to me there's a ground issue somewhere.
Check and confirm that...
Freya:
- remove the lid and see how they coupled the:
a) Mains IEC ground to chassis. Is there a green/yellow wire connecting the mains ground and Freya's chassis?
b) Freya's PCBs to chassis... DC-coupled or AC-coupled (using capacitors).
c) How are the XLR outs connected to the chassis? ... they shouldn't be.
If you do not want to open Freya.... just send an e-mail to Schiit, explain your problem, ask the last 3 questions (a,b, and c).. and see what the answer is.
Good luck,
Nick
Check and confirm that...
- Each XLR shield pin is connected to the GND input on the corresponding AMP PCB
- Aleph J AMP PCBs have a single ground wire/connection back to PS PCB. E.g. each APM PCB will have a single ground wire going back o the PS PCB.
- The PS PCB has a single point of connection to the chassis. Here, you could use the 10 ohm NTC (CL60) between PS PCB ground and chassis.
- Mains IEC connector ground lug is connected to the chassis.
- Spekers' negative wires are connected to the PS PCB ground.
Freya:
- remove the lid and see how they coupled the:
a) Mains IEC ground to chassis. Is there a green/yellow wire connecting the mains ground and Freya's chassis?
b) Freya's PCBs to chassis... DC-coupled or AC-coupled (using capacitors).
c) How are the XLR outs connected to the chassis? ... they shouldn't be.
If you do not want to open Freya.... just send an e-mail to Schiit, explain your problem, ask the last 3 questions (a,b, and c).. and see what the answer is.
Good luck,
Nick
Ok... I had a look at your last few posts, and at what I wrote just above... and, I think you may have the following situation. (of course, I might be completely wrong here.. 
)
The PS PCB you use in Aleph J, may not be grounded to the chassis. So, none of your AMP PCBs is grounded to the chassis... So, there is no connection between the PCBs ground, and the Alep J IEC connector ground lug (i.e. - earth).
When you connect the Freya, the ground return path is like this:
Alep J: the PS PCB ground return will go back to the AMP PCBs, and then back using the XLR common... all the way (via interconnects shield).. to Freay... and will eventually hit the Freya's IEC ground lug.
As I said... I might be completely wrong here... and that's why I asked you to check a few things... in my previous post
)The PS PCB you use in Aleph J, may not be grounded to the chassis. So, none of your AMP PCBs is grounded to the chassis... So, there is no connection between the PCBs ground, and the Alep J IEC connector ground lug (i.e. - earth).
When you connect the Freya, the ground return path is like this:
Alep J: the PS PCB ground return will go back to the AMP PCBs, and then back using the XLR common... all the way (via interconnects shield).. to Freay... and will eventually hit the Freya's IEC ground lug.
As I said... I might be completely wrong here... and that's why I asked you to check a few things... in my previous post
Thanks for all of your suggestions and assistance, Nick.
You are correct - the only connection from Aleph PSU ground and Aleph chassis is through the thermistor to the star ground.
On the Aleph, I measure 0 ohms between: PSU ground; negative speaker posts; both XLR pin 1; and amp board ground - consistent with your recommendation earlier in the thread. IEC ground is connected directly to chassis (star ground).
HOWEVER, when measuring resistance between PSU ground and chassis, the value appears to start in the hundreds of ohms and steadily increases - as though the meter is charging a cap... This is not new, and has bugged me since I removed the short between the PSU ground and chassis. I measure a steady 33 ohms across the thermistor between star ground and chassis.
I am presently checking for a potential short between a cap on the board and chassis ground.
In regard to the Freya, there is continuity (0 ohms) between:
Thanks again,
Joe
You are correct - the only connection from Aleph PSU ground and Aleph chassis is through the thermistor to the star ground.
On the Aleph, I measure 0 ohms between: PSU ground; negative speaker posts; both XLR pin 1; and amp board ground - consistent with your recommendation earlier in the thread. IEC ground is connected directly to chassis (star ground).
HOWEVER, when measuring resistance between PSU ground and chassis, the value appears to start in the hundreds of ohms and steadily increases - as though the meter is charging a cap... This is not new, and has bugged me since I removed the short between the PSU ground and chassis. I measure a steady 33 ohms across the thermistor between star ground and chassis.
I am presently checking for a potential short between a cap on the board and chassis ground.
In regard to the Freya, there is continuity (0 ohms) between:
- IEC ground
- Pin 1 of XLR In
- Pin 1 of XLR Out
- RCA Out-
Thanks again,
Joe
It seems you are on the right track.
"HOWEVER, when measuring resistance between PSU ground and chassis, the value appears to start in the hundreds of ohms and steadily increases - as though the meter is charging a cap..."
The above is not right... The PS GND must be DC coupled (either via direct link or via a CL60/ diode bridge... what not...used widely on these threads), to the Aleph J chassis. So, there's something dodgy going on with the PS PCB... it seems.
"
The above seems to be okay...
"HOWEVER, when measuring resistance between PSU ground and chassis, the value appears to start in the hundreds of ohms and steadily increases - as though the meter is charging a cap..."
The above is not right... The PS GND must be DC coupled (either via direct link or via a CL60/ diode bridge... what not...used widely on these threads), to the Aleph J chassis. So, there's something dodgy going on with the PS PCB... it seems.
"
- IEC ground
- Pin 1 of XLR In
- Pin 1 of XLR Out
- RCA Out-
The above seems to be okay...
I have been using the same PSU PCB (BrianGT / Chipamp site) with my mini-Aleph (also made on BrianGT boards). I made a connection between PSU ground and chassis / IEC PE via a CL-60.
So, in principle, this PSU PCB is known to be able to work in an Aleph ...
Best regards, Claas
So, in principle, this PSU PCB is known to be able to work in an Aleph ...
Best regards, Claas
Not all grounds are created equal…
I had tied the negative rail ground, rather than the common PSU ground, to the thermistor.
So it was, indeed, charging a cap - three of them, actually!
While I hesitate to call the Aleph J project complete, at least Freya and J are on speaking terms and maybe I can relax and enjoy some music this weekend.
With TWO boneheaded power supply mistakes, how this thing ever worked at all is beyond me.
Thanks again for everyone’s input, but just as importantly thanks to the diyAudio community at large for encouraging me to THINK.
I had tied the negative rail ground, rather than the common PSU ground, to the thermistor.
So it was, indeed, charging a cap - three of them, actually!
While I hesitate to call the Aleph J project complete, at least Freya and J are on speaking terms and maybe I can relax and enjoy some music this weekend.
With TWO boneheaded power supply mistakes, how this thing ever worked at all is beyond me.
Thanks again for everyone’s input, but just as importantly thanks to the diyAudio community at large for encouraging me to THINK.
Perhaps that speaks to how resilient this design is. It won't be happy driving a 1 ohm load, but it will do so without blowing up. "Stable into a dead short", Nelson once said, IIRC.
I once hooked up the power supply backwards on a B1 and blew the filter caps.BTW, you're not the first person to have the ground thermistor in the wrong place. I recall someone doing the same with an F5 and unfortunately zapped an upstream component.