or Connect
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › A Very Compact Hybrid Amp
New Posts  All Forums:Forum Nav:

A Very Compact Hybrid Amp - Page 85

post #1261 of 2213
check the emitters of q4p and q3p and the power is perfectly clean.

changed all input wiring to shielded wire.

checked the 24v power rail, again perfectly clean.


noise is present with pot all the way down....

looks like this has to be noise on the ground.
post #1262 of 2213
Thread Starter 
Been having a background conversation with Adamus as he noted and he has been taking some measurements. I had a theory of a potential problem but have discarded it after thinking it through.

What adamus has done is to scope all of the rail voltages. They are all squeaky clean except for the 40mV on the heater. He has also placed a grounded tube shield on the tube. Nothing changes.

So, with grids grounded, tube shield on, and clean voltages the only other possible candidate is a ground problem.

This might take a little thought . . .
Dr. Cavalli gained notoriety with his first DIY amplifier projects. His success has blossomed into Cavalli Audio, a world leader in amplifier design.
Reply
post #1263 of 2213
ok, i think we have a breakthrough.

I cut the pins off a 6n1p that went bad, only pins 4 and 5 remain.

with the heaters on there is noise, pulling the tube out and the noise disappears.
post #1264 of 2213
Thread Starter 
Really appreciate the tests that Adamus has been running and this is a real breakthrough on this problem.

What is says is that the noise appears when the heater supply is loaded whether the tube is there or not. So EM radiation or other tube noise is not the issue.

This means that loading the heater supply is affecting the other supplies somehow.

We are working on it . . .
Dr. Cavalli gained notoriety with his first DIY amplifier projects. His success has blossomed into Cavalli Audio, a world leader in amplifier design.
Reply
post #1265 of 2213
Quote:
Originally Posted by runeight View Post
... He has also placed a grounded tube shield on the tube. ..
Forgot to report I also tried this last night - same result. With my upgraded caps, i only have 7-9mV on the heaters.

Looking forward to...something! Let me know if you need another guinea pig.
post #1266 of 2213
Thread Starter 
Why yes, while Adamus is sleeping ...

What we want to investigate is what happens to the LV supply when the heaters are turned on and off.

If you have a 22R 5W resistor you can stick this into pins 4&5. To simulate a 6H30 use 10R. Both at the 6.3V setting.

Or you can do what Adamus did and take a bad tube and cut all of the pins off except for 4&5.

Do you also hear a change in noise when the heater circuit is loaded and when it is not? If so . . .

If you have a scope we need to see what the voltage looks like both before and after the regulator. DC and AC components.

Since the tube is not in the circuit it can't be the B+ supply. It can only be the LV supply or a grounding issue.
Dr. Cavalli gained notoriety with his first DIY amplifier projects. His success has blossomed into Cavalli Audio, a world leader in amplifier design.
Reply
post #1267 of 2213
Quote:
Originally Posted by adamus View Post
ok, i think we have a breakthrough.
I cut the pins off a 6n1p that went bad, only pins 4 and 5 remain.
with the heaters on there is noise, pulling the tube out and the noise disappears.
My testing does not support the above unless qualifiers are added ( 5mv noise w/loaded heater and no tube, upto 30mv noise w/tube).

Here's what I did:
  • Shorted R18s (output resistors) to hear noise better. W/6BZ7 noise is very apparent now w/D2000s.
  • Using a "bad" 6BZ7, cut off all but pins 4/5 and tested. Noise was much less audible & only averaged 5mv on my scope.
  • Decided to really load switcher. Used 10R 10W resistor across heater pins. This draws about 600ma @ 6.3v. Same results as tube w/pins cut-off - Noise much, much lower. In fact, not above ambient noise level on my scope, about 5mv, but of different content than ambient noise.

So although heater circuit plays a role in this noise, I do not believe it comes down to it simply supplying load current/wattage.

If you do resistor load tests watch out you do not exceed 800ma draw. E.g. at one point I used a 15R resistor in load testing. At 6.3v it drew 360ma, at 12.6v it draw about 800ma & the heater circuit made an audible buzzing sound.

So more research to do. This may wind up being 1) switcher causing ground contamination (isolate its ground/ground plane?), 2) switcher causing back-EMI into common A/C supply, 3) a bit of both. Note we are not employing a line-in EMI filter here (typically chokes & capacitors) and it looks like that might require a dedicated 1" X 2" PCB area that we do not have.
post #1268 of 2213
Quote:
Originally Posted by runeight View Post
Why yes, while Adamus is sleeping ...

What we want to investigate is what happens to the LV supply when the heaters are turned on and off.
Sorry i missed this, went to sleepy...


Let's see where we're at tonight, scope is on the bench. I may need some help on using it (Where to attach the ground lead SG or closer to the measurement point (Pin 4?)
post #1269 of 2213
Just an FYI: I'm doing my noise measurements at the amp's outputs... This is what really counts as this is what we hear. Again, think about shorting R18s to make the noise more audible.

I'll be researching more tonight & tomorrow as well. And since the only way I can get significant noise to come out of the amp is with a 6.3v heated tube (with all its pins), that's the way I'll be testing.

I'll be looking to see about isolating the switcher's ground plane (incls pulling L1H away from main amp's plane), maybe raising both heater's coils well off the PCB, maybe pulling D1H + D3H & feeding switcher from another 24V rectified supply, maybe fooling around with physical shielding for switcher as a whole (bent up sheet metal).

All of this to find if noise is made up of 1) (common) ground plane contamination, 2) general EMI broadcasting, 3) backfeed EMI into common A/C (that HV & LV are not filtering). May be some combo. But what do I know So others can pursue other ideas.

Edit: A little note WRT my "we are not employing a line-in EMI filter here " for SM circuit. I don't think one is needed for backfeed EMI to mains/AC as the wallwart is providing this isolation. Only a consideration if backfeed EMI is found the primary cause of our noise.
post #1270 of 2213
Quote:
Originally Posted by cfcubed View Post
I'll be looking to see about isolating the switcher's ground plane (incls pulling L1H away from main amp's plane),
My proto currently has both coils mounted up at least 1/2" to 3/4", That's the only way the 1 amp versions fit. didn't notice a differenc in noise level, but perhaps some combo. Listening now with my Tung Sol 12Au7, nice and quiet with the RS-1s . but we know that
post #1271 of 2213
hows the testing going chaps? I am all tested out. I find spikes on the 24v line with a tube in that i dont find without.

If anyone has a scope, please have a go at confirming my results.

Cheers
post #1272 of 2213
Thread Starter 
Remember that the heater ground plane is already isolated from the main ground plain. They only contact at one point near the SG. If the SG is well grounded no heater ground noise should sneak over to the main ground plane.

Look carefully at your board ground planes and you will see this.
Dr. Cavalli gained notoriety with his first DIY amplifier projects. His success has blossomed into Cavalli Audio, a world leader in amplifier design.
Reply
post #1273 of 2213
I believe heater's L1H is outside the heater ground plane, but don't think this matters anyway.

With disabled heater circuit & battery heater no hum is heard w/6.3v heated tube. A good deal of hiss/white noise is heard w/R18s shorted, battery heater & D2000s. But this is all about the hum style of noise.

So somehow the heater circuit is causing this issue. More research is needed. Once again, issue is largely limited to use of non-12AU7s & sensitive cans & is heard when there is no input signal.
post #1274 of 2213
Quote:
Originally Posted by adamus View Post
hows the testing going chaps? I am all tested out. I find spikes on the 24v line with a tube in that i dont find without.

If anyone has a scope, please have a go at confirming my results.

Cheers

Scopes are Kewl!

Ok, Alex, I'll systematically respond to your questions - but I tried this first. As reproted my AC mV on the Heater supply are ~5 (with all my extra capacitance)

Phones 32ohm PX100 100R R18

AC mV on 24V supply (probe on Vout, SG):
DC is right at 24V

12V No Tube .3-2.0mV very very slight Hum must press phones to head)
12V 12AU7 .3-2.0mV Very Slight Hum

6V No Tube .3-2.0mV very very slight Hum

6V 6BQ7 heater pins only 2.8-3.x mV very very slight Hum

6V 6H23 1.7-2.3mV slight hum

6V 6H30 6-12mV slight hum

Hum is mostly in right ear.

Any help, or adding to the confusion?

EDIT: Even with the 6H30 hum is very low tonight.
post #1275 of 2213
Even with a 6N1P is very slight with the px100, have to press senn 650s to hear any hum at all. I swear this was noisier Sunday night.

Cosmic Rays?

I'm gonna case her back up and see...
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › A Very Compact Hybrid Amp