Need Help with First Tube Build - WE91 300B Parafeed Derivative

[Thermioniclife]

Glad to hear that you got it sorted out.
Best wishes to you and yours.
Regards,
Lee

[EricS]

After a long time not being able to work on my amp, I have been able to make some progress again.  I was experimenting with positioning the RCA input next to the input tube (away from the AC mains input and speaker output) vs putting the RCA input next to the output transformer and speaker output terminals.  There is very minimal difference in speaker output noise levels, both measure mostly less than 1.0mV AC noise on my DMM.

I think I prefer having all of the inputs/outputs in a line along a single edge of the amp because it makes routing wires in and out of the amp visually more neat on the equipment rack.  Is there any reason that I should not do this? 

With either RCA input placement, the amp sounds VERY quiet on my Tang Band 1772 speakers (advertised sensitive of 95dB - this is probably optimistic after my correction network to tame the treble).  I have attached an image of each RCA input configuration, as well as an image of the output noise level on my scope with the RCA input "open" (not shorted).  The scope is set to display 5mV per division on the y-axis with a time window of 2ms per division on the x-axis.

Thanks for any insights.

[Thermioniclife]

You shouldn't have any problems with the input and output next to each other.
I did this on a pair of 300b mono's that I built a while back.

[EricS]

Thanks, Lee.  I've been playing around with the prototype top plate and I'm happy enough to drill holes anywhere, but I'm becoming a bit more conservative as I'm thinking about the work required to make two final plates and get them powder coated.

[EricS]

Work and projects around the house have been keeping me pretty busy lately, so my progress has still been slow.  But, I'm coming down the home stretch.  Got my aluminum back from the powder coating shop last week and thought it turned out very nice.  I went with a textured black (thanks for the suggestion, PB!) with speckles of gold in it for the base plate and a gold-ish color for the power transformer bells in an effort to come close to the brass bells of the output transformer.  Both were a dual-layer powder coat process.  I also stripped the shrink wrap from the PSU caps and painted them with a gloss black enamel so they wouldn't have silver circles on top of them.

I had an interesting problem with the aluminum plates.  The sheets that I received from the shop were slightly bowed, so I took them back and asked them to run the sheets through their rollers to flatten them out.  A few minutes later, I had nice flat sheets in my hands.  But when I picked them up from the powder coat shop, the bow had returned - I guess the heat from the powder coating process caused them to spring again.  Thus, there is a slight bow to the top plate that flattens out a little with the weight of all of the components on the plate.   Guess I could always mount an "L" bracket to the inside of the wooden chassis and drive a screw through the top plate into the L to pull it flat again.  I think I'll leave it alone for now...

I've attached a few images of the top of the amp.  I haven't finished the wiring yet, so the amp won't be functional for another few days.   

[Paul Birkeland]

The bow will go away over time and comes from the production of the sheets themselves.

[Deke609]

Eric - It looks amazing! Nicely done.  Happy to see a new post about your progress. Looking forward to the finish and listening impressions.

cheers, Derek

[EricS]

Good to know that the bow will go away over time.  I was a bit disappointed to see it return after the heat curing of the powder coat process.

Can't wait to hear the stereo pair of them!

[Thermioniclife]

Nice job Eric, Sweet inlay and finish. Something to be proud of.

[EricS]

As I'm nearing completion, I have a question about using/setting the hum pot in this design.  My DC filament supply for the 300B is a CLCR (4,700uF, Hammond 155B, 10,000uF, 0R5 10W).  The hum pot is a 15R 5w pot flanked by 22R 5w resistors.   AC ripple on the 300B heater is essentially not measurable on my DMM (0.0mV).  As a result, adjusting the hum pot makes no audible difference at the speaker output (I'm using Tang Band 1772 drivers right now somewhere near 92-93dB efficiency after I tame the treble). 

Given that I can't *hear* any difference, where should I set the pot for best/safest operation?  Should I just leave it in the middle of the range?

Thanks for any insights!

[Paul Birkeland]

I would just put it in the middle.  You may be able to measure a null with a scope or FFT, but if you can't hear it I wouldn't bother!

[EricS]

Thanks for confirming, Paul!

[EricS]

Happy New Year!

Wow - I didn't realize it's been since July that I've worked on my amps!  The fall semester kept me pretty busy, but now I've got just a little bit of time on my hands again and I'm hoping to get these amps finalized.

I may have made a mistake in the context of my mental health, but I recently picked up a nice USB Analog to Digital Converter so I could more easily do some sound/noise measurements.  Maybe there is something to that old "ignorance is bliss" sort of thing ;-)

Anyhow, I measured the noise floor for my prototype amp.  The graph below is what I am measuring with the input shorted and an 8R load on the output.  Removing the input shorting cap at the RCA in makes no difference in the noise profile.   But, since I have no context, I'm wondering what the noise floor "should" look like for an amp like this.  Is this a good result, or do I still have some work to do? 

To me, it looks like I have some power supply noise to deal with, as evidenced by the peaks at 60Hz, 120Hz, 180Hz, 240Hz, and 300Hz.   The odd harmonics seem to be a bit higher in magnitude than the even harmonics.  Measurements at the speaker terminal indicate about 42mVDC offset and 1.3mVAC hum.  The noise floor on the ADC all by itself is about -120dB.

Any input would be appreciated! 

[Paul Birkeland]

1.3mV of hum would be OK with me.  That's not quite as low as it could be with a DC supply, but about 10dB quieter than an AC heated amp. 

The FFT image you posted is useful in some ways, but not necessarily as useful as knowing the actual voltage. 

Incidentally you have two power supplies that will make 120Hz noise, not 60Hz noise.  The 60Hz noise you're seeing could be a ground loop issue or magnetic coupling between the PT and your output iron.

[Deke609]

Hey Eric! Welcome back! I was wondering what happened to you and this amp.

Could some of that noise be residual power supply ripple appearing at 300B grid via the plate of the driver tube? If so, maybe consider a cascode constant source as driver load?

just a thought.

cheers, Derek