Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
With a 1V amplitude unbalanced input signal at a frequency of 40Hz or less, my Sympaticos are motor boating. The gain is set at 26dB. The speaker (sub) DC resistance is around 6 ohms. Is this normal? Any suggestions on what I should do? Please help, I'm about to give up. Edited by user Saturday, May 19, 2012 3:20:43 PM(UTC)
| Reason: changed 6dB to 6 ohms
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
Here is a scope capture of what is happening. The sound is a fairly loud and dry repetitive pop (like an engine running). https://plus.google.com/...?authkey=CIie-Z37rPys4gEI would really welcome Russ' insight on this. I really don't know what to do next. Regards, mcoteca attached the following image(s):  IMG-20120518-00260.jpg (196kb) downloaded 17 time(s).You cannot view/download attachments. Try to login or register.
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 2,868
Location: Massachusetts, USA
Thanks: 2 times
Was thanked: 141 time(s) in 134 post(s)
|
Can you post a picture of your build? It helps answer a bunch of questions in one go.
I am in training for a couple days and Russ is dealing with some personal matters, but we will get you sorted.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
Hi Brian, thank you for your reply. Links to pictures and further information provided below. Feel free to get back to me when you have time. Rgds, mc Pics of my Sympatico boards build (actually, I'm only showing one): https://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkYou will see that the yellow pad is installed to isolate the heat sink from the LM4780. Voltmeter measurements seem to confirm that the heat sink is isolated; however, I can't help but wonder if there could be a near-DC sneak path that would explain my problem, maybe the screw is threading perilously close to the LM4780 package? Housing assembly pics, consisting of two Sympaticos integrated with a subwoofer active crossover from Linkwitz Labs (which I also assembled): https://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkIn my application, the Sympaticos are used for driving a pair of subwoofers. I tried to calculate the maximum voltage they will manage on my setup, but I'm not sure what is the current limit (closer to 5A or closer to 10A?) Further, would the feedback loop of the amp by any chance be designed/optimized to achieve stability with a load impedance no less than 8 ohms? The DC impedance of my subs is around 6ohms. To isolate the problem at hand, the other electronics in my housing were left unpowered as I have implemented two independently switched 110V circuits, one of them dedicated to the Sympaticos and their associated mains transformers, Anteks AN-4222; these guys are heavy!). Note that I'm also chasing a 200mV DC offset in the subwoofer crossover electronics assembly, which I'm investigating as a separate issue. Therefore, the scope capture provided in my earlier post was obtained from the right channel amp fed directly from a high quality commercial DAC with volume control. Signal ground and chassis ground are isolated (again, I wonder if I could have an undetected near DC sneak path here that could cause my issue; I tried with the chassis both connected and disconnected from earth ground with same results). Edited by user Saturday, May 19, 2012 3:14:08 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 2,868
Location: Massachusetts, USA
Thanks: 2 times
Was thanked: 141 time(s) in 134 post(s)
|
Are you tying -IN to GND?
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
This sound like a ground loop of some sort, try completely isolating the input and the speakers. What I mean is do not connect to the case. Also you absolutely need to be sure your screws are properly isolated from the heat sink too.
4-Ohm load is no problem at all.
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
Show us a picture once you have it wired up.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
Hello Russ, I believe that everything has been isolated, but perhaps you will have some comments on my set-up. Here is what I did : - disconnected everything from the right channel amp that was not essential (for example, I disconnected the front panel extra LEDs) - chassis disconnected from earth ground - double checked that signal ground is isolated from the chassis - removed LM4780 screws from the housing heat sink; I need to keep my tests short since this is degrading the thermal path - added a second heat pad underneath the first one to provide increased electrical isolation; I'm now absolutely sure that the LM4780 body is isolated from the heat sink - used a different cable to the speaker (just in case) - the input comes directly from my off-the-shelf DAC - power comes from dedicated AN-4222 trafo connected to AC1, AC2, CT1, CT2 - note the Sympatico feet are still bolted to the chassis, but my understanding is that the PCB mounting pads are isolated The results are the same, unfortunately; it starts acting up at -10dB, i.e.1V, 30Hz pure tone. Perhaps the next step should be to remove everything from the housing and try again? I'll wait for your feedback first. So far I have concentrated my efforts on the right channel amp; however, my left channel amp has the same problem. Pictures: https://picasaweb.google...0liipFm0?feat=directlinkhttps://picasaweb.google...0liipFm0?feat=directlinkRegards, Martin Edited by user Sunday, May 20, 2012 4:11:36 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 12/24/2008(UTC)
Posts: 100
Location: Seattle, WA
|
I looked at the other pictures in your Pluto library...
Have you checked that you get the same problem with a different isolated source? It might be worthwhile to try that before taking things further apart. A cell phone or ipod will suffice for that. Do you get the same behavior on both channels?
My Sympatico driven balanced from a Buffalo II/IVY III is dead slient in a similar dual mono setup.
Cheers
Thomas
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
Hi Thomas,
the Pluto was my first DIY project; there is indeed a collection of pictures of that project on my Picasa page, which was absolutely great fun to do (the project, not the pictures); I'm really happy how the speakers turned out. I thought long and hard about doing the Buffalo for my second project, but I haven't taken the plunge yet. In the mean time, I decided to do the subs project for the Plutos which require a dedicated stereo amp, hence the Sympaticos.
But I digress... Yes, I tried with my portable mp3 player (BlackBerry phone). The problem is still there, using a pure tone at 20Hz, around 0.4V RMS (which is the maximum voltage the player can output). I then tried at 60Hz (where I had a bit more signal voltage incidentally, around 0.5V RMS); sound was 100% clean.
Thanks,
Martin
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
I don't see where -IN and IG at the Sympatico terminal block are shorted together as they should be. Are they? Edited by user Sunday, May 20, 2012 11:08:28 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
Also - do not run without the LM4780 securely connected to the Heat Sink - or you will kill the amp.
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 2,868
Location: Massachusetts, USA
Thanks: 2 times
Was thanked: 141 time(s) in 134 post(s)
|
Two layers of thermal pad is not a good idea either. One pad is sufficient.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
Hello, here are some answers:
- The unbalanced input ground is y-split into ground and IN-. It's hard to see in the picture.
- I'll put the screws back in, just wanted to do some quick experiments to rule out any possible short to the heat sink.
- The second heat pad was part of the experiment. I will go back to a single heat pad.
Regards,
Martin
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
Hello, one thing I plan to do shortly is feeding a balanced input signal to the amp and see what happens. My commercial DAC has both balanced and unbalanced outputs so it should be easy to give it a shot. I just need to get my hands on an xlr female connector and jury rig a cable. If this clears the problem, then one permanent solution would be to build a single-to-differential stage with a couple of op amps. But not knowing what is currently going on, this is just a shot in the dark, so to speak.
mc
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
There is no need for a SE/BAL stage. The amp itself is one.
I use the amp regularly with SE inputs.
I suspect something completely different.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
Hi Russ, in that case I won't proceed with this. Let me know if there's anything else I could do or if you prefer I put things on hold for now. My offer to take a trafo and one of the amps out of the housing is still there, although I'm quite confident that the last test I did achieved the same thing within the housing. But who knows. I'm clueless as of now. It's gone beyond my level of understanding. Best Regards, Martin Edited by user Tuesday, May 22, 2012 9:08:01 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 10/11/2011(UTC)
Posts: 40
Location: Montreal, Canada
Thanks: 1 times
|
PM sent to Russ. Would love to hear back from Brian or Russ.
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
Try bridging -IN and GND right at the terminal. Solder a part of a resistor lead across the pads on the bottom.
After that please very carefully measure the voltage across all resistors.
|
|
|
|
|
|
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.
