L0rdGwyn's DIY Audio
Jun 8, 2020 at 7:51 PM Post #466 of 6,723
Made some progress today on the beast.

Heat sink is all prepped and ready for regulators, holes are drilled for three Maidas and four Rod Colemans.

IMAG2139.jpg

Mounted the Hammond filament transformers and raw DC supplies, should be able to get the filament supply up and running tomorrow, then testing of the B+ supplies will begin.

IMAG2143.jpg

I should build all amps on plywood! Low stress, no worries about precise measurements, comes together quickly too :ksc75smile:
 
Last edited:
Jun 9, 2020 at 6:54 PM Post #467 of 6,723
Made some solid progress today.

IMAG2147.jpg

Got all of the regulator boards completed and mounted to the Heat Sink. Maidas are all ready to go for when I get to work on the B+/bias supplies.

IMAG2146.jpg

Tube sockets were all mounted, wired up the filament supply and got the filament voltages dialed in. This was my goal for the day, happy I got to see the tube glow :)

IMAG2158-2.jpg IMAG2149.jpg

Next up will be the B+, source follower, and bias supplies. Will get started on them this weekend, going to be a lot of work. Just gonna take it one power supply at a time...
 
Last edited:
Jun 12, 2020 at 10:14 AM Post #468 of 6,723
The eagle has landed, the Omega Super 8 High-Output Monitors have arrived. The veneer is figured walnut.

IMAG2165-2.jpg

They were waiting for me when I got home from work last night, of course I threw them in the stereo and gave them a quick listen despite being completely exhausted, more listening this morning and throughout the day. Omegas are notorious for their long break-in period (or brain-in?), so I will defer giving major impressions until later, maybe past the 100-hour mark. I will say this right off the bat - the midrange detail and vocal nuance is impressive. They are also much more sensitive to on-axis positioning than my Snells. From my short positional experiments, I've found they need a pretty generous toe-in. Their sound is also more sensitive to changing tube configurations than the Snells.

IMAG2164-2.jpg

From a DIY standpoint, I can now do Snell surgery. Will have to decide where this project falls in terms of priority with the 841/801A amplifier prototype underway, but I plan to re-cap and re-resistor the crossovers and replace the tweeter ferrofluid.
 
Last edited:
Jun 12, 2020 at 12:28 PM Post #469 of 6,723
Happy new speaker day. :)

You'll want to experiment with toe-in a lot. They're sensitive to that, and also to their distance from the back wall. Louis seemed to have incorporated the rear port tuning into the voicing of his speakers because they change character a lot depending upon their proximity to walls. Your position relative to them (being on, inside, or outside the imaginary "triangle" you form with their position) will affect the treble response dramatically. The high end rolls off with increasing angle. The aperture of the whizzer cone is actually a pretty good guide. If you can't see the inner edge of the inside of the whizzer you're too close to them or don't have them toed enough. I find the halfway point between that and havintg them pointed directly at you to be a nice sweet spot for treble balance, but my Alnicos might be voiced a bit differently too.
 
Jun 12, 2020 at 1:07 PM Post #470 of 6,723
Thanks, @Xcalibur255 . Yes, definitely will be a period of experimentation, I began working on the toe-in and wall distance last night. Unfortunately, due to my listening space being lived-in, I am pretty limited to how far into the room I can place the speakers, we will see. I had expected a directly on-axis toe-in to be too aggressive, surprisingly I didn't find that to be the case, but right now I have them slightly less that direct, just outside the ears. I'm sure that will be adjusted with break-in/brain-in. I may also rewire the OPT of my SET - currently wired for an 8ohm secondary, whereas the Super 8 HO have a 4-6ohm nominal impedance, so damping factor is not optimal. In the long run, my plan is to pair the 801A amplifier with the Snells and the 6A5(?) amplifier with the Omegas, due to the differences in sensitivity and power output of the amplifiers.

I got another DIY surprise today, the first of the R120 pairs has arrived.

IMAG2166-2.jpg

This puts the overhaul of my 6A5 amp on the table too, three simultaneous projects. Whoof! I am taking it easy today and listening to the Omegas, will decide which path to go down tomorrow.
 
Jun 12, 2020 at 2:53 PM Post #471 of 6,723
The top end will wake up in a few hours and you might find yourself dialing your toe-in back a bit in response to that. I personally like to sit quite close to speakers and be "inside the triangle" so I toe-in more aggressively to compensate for that and maintain imaging stability. This listening style makes the whole front stereo image a bit less precise but in exchange it expands field width and depth and increases immersion which I find a worthwhile trade-off. I think what I do actually makes for a less accurate presentation, but it's what I like. When you position yourself in the 'traditional' way (equilateral triangle between you and them), and do a half toe the stereo image is holographic and you identify instrument positions down to the millimeter in space. Since I do nearfield listening anyway I trade a bit of that off to get more of a "there's sound everywhere" feeling instead.

In some ways the speakers can feel a bit fussy about placement, but they are also pretty forgiving in some ways too and will accommodate more than once preference if you tinker around with it.
 
Jun 12, 2020 at 3:53 PM Post #472 of 6,723
@Xcalibur255 I'll keep your notes in mind as the sound evolves! I do have to wonder if the same rules apply between the alnico and ferrite drivers. Considering they are from the same designer, I would think so but can't be certain, I will keep you posted. At the time of my purchase, I was looking for something of a "well-rounded" speaker that could do well with most genres, I have a pretty eclectic taste in music. Had I known I was going to land a pair of very nice Snell Js, which I believe fall into that category, I probably would have gone with the Super Alnico Monitor from Omega, I just wasn't sure at the time that it was going to be able to do everything I wanted. Well hindsight is 20/20, something to keep in mind for the future :wink:

Plans for my day changed quite dramatically and I got some time back. I am going to do some critical listening with the 6A5G, then make the minor modifications to use the R120, see if I can make a decision on whether or not I will be moving forward with them.
 
Jun 12, 2020 at 5:08 PM Post #473 of 6,723
Here is the R120 in situ. Fivre 5Z3 rectifier to drop the B+ to a happy place relative to the U18/20.

DSCF5700-2.jpg DSCF5696.jpg

Yes, these are very nice sounding tubes. Sound is a little more intimate than the 6A5G, better bass definition, a little more forward presentation overall. The 6A5G tends toward smooth and laid back to my ears. Could be less distortion, will have to take some measurements to be sure. The additional gain of the R120 is noticeable on the volume knob, these tubes give me the flexibility to make some other modifications to the amp, so it is something of a multi-faceted decision. I think this operation is a go, but going to give it some more listening time.
 
Jun 12, 2020 at 9:34 PM Post #474 of 6,723
Been listening to the Omegas with the R120 for a few hours now. Can't say how much is burn-in and how much is brain-in, but the sound has changed, for the better.

To be honest, on first listen last night, the speakers sounded pretty congested and the imaging was not great, really felt like I was listening to two point sources. Granted listening after a long work shift never yields good results for me and I had Omegas really do need some break-in, so I slept on it.

Throughout the day, the imaging has drastically improved, much more cohesive and larger staging. The treble has "woken up", as @Xcalibur255 suggested, I towed the speakers out a bit more as a result. The bass seems to have tightened up and deepened a bit as well, but oddly enough the Snells seems to reach lower from memory and are more "full" sounding, which was unexpected, we will see how it continues to progress.

As far as strengths, right now they seem lie in the upper midrange and treble. It just "pops", very realistic, nuanced, the human voice is extremely textured and lifelike. Anything with female vocals or a male falsetto is electrifying.

The presentation of these speakers is very different than the Snells, will take some time to put it into words, looking forward to a more direct A-B comparison with the Snells once I'm done messing around with the crossovers. A different flavor for sure, and I am really enjoy it :)

DSCF5706.jpg
 
Last edited:
Jun 13, 2020 at 8:47 AM Post #475 of 6,723
An interesting note about the R120.

It's pretty obvious from the external construction of the tube that it is a variant of the European-made ELXX series, most similar in construction to the Philips EL39 from what I have seen, it is a pentode converted to triode. Taking a look inside, we can see there is a screen grid in place that is connected to the plate. While there is no suppressor grid, the rods for g3 are intact and connected internally to the cathode.

IMAG2176.jpg

So the R120 is a former pentode internally strapped to a triode, but without a suppressor grid. What an oddity!

I am going to put together the first of the regulated supplies for the 841/801A amplifier today, most likely the source follower V+ at 150V.
 
Last edited:
Jun 13, 2020 at 8:27 PM Post #476 of 6,723
It's a bit of a mess, but two of the four power supplies are up and running.

IMAG2198.jpg

This dual secondary 150V Antek transformer feeds both the source follower V+ and source follower V- / bias supplies.

Just to recap, this amplifier will be using a transistor buffer stage between the input and output tubes. Think of it as the transistor version of a cathode follower - unity gain, low distortion, and low output impedance, supplying enough current to drive the grid of the 801A +40V while drawing ~10mA at maximum positive swing.

The first of the two supplies is the 150V input to the source follower drain. It will supply the grid current drawn by the 801A on positive grid excursions and is Maida regulated. With the way I have parted the Maida, it is adjustable from 140-160V with a 1-turn pot on the board, perfect.

The second supply is an adaptation of Pete Millett's shunt-regulated bias supply. I had to build his circuit in LTSpice and repart it for my purposes as his was original built for 300B bias with little-to-no current draw. It is adjustable from -150-160V, it will feed the CCS providing bias current on the source of the source follower, as well as feeding the fixed bias supply of the 801A.

IMAG2203.jpg

Everything looks good with the dummies, both supplies dialed in to +/- 150V. Tomorrow I will get started on the big fellas, the 450V B+ supply for the 801A and the 650V B+ supply for the 841. Watch your fingers and toes :)
 
Last edited:
Jun 14, 2020 at 12:00 PM Post #477 of 6,723
841 B+ supply is up, 650V Maida regulated voltage doubler. This supply will have a low current draw from the 841 CCS load, around 18mA, using a voltage doubler allowed me to use a relatively small 250V toroid.

Nothing went up in smoke went brought up on variac, which tells me parts were well-selected :)

IMAG2205-2.jpg

Gonna take a break then throw together the 801A B+ supply. If I have time today, will see if I can get my B+ time delay circuits working, I picked up some kits from K&K Audio. They can't take the high voltage of the 801A / 841 B+, but because these are dedicated B+ supplies, they should work on the mains transformer primaries without issue. The boards will be powered by the 6.3VAC windings of the 150V source follower mains transformer, which will not have a time delay. The 30 second time delay in addition to the built-in soft start of the Maida regulators will give the filaments more than enough time to come up to voltage before B+ is applied.
 
Jun 14, 2020 at 1:37 PM Post #478 of 6,723
841 B+ supply is up, 650V Maida regulated voltage doubler. This supply will have a low current draw from the 841 CCS load, around 18mA, using a voltage doubler allowed me to use a relatively small 250V toroid.

Nothing went up in smoke went brought up on variac, which tells me parts were well-selected :)

IMAG2205-2.jpg

Gonna take a break then throw together the 801A B+ supply. If I have time today, will see if I can get my B+ time delay circuits working, I picked up some kits from K&K Audio. They can't take the high voltage of the 801A / 841 B+, but because these are dedicated B+ supplies, they should work on the mains transformer primaries without issue. The boards will be powered by the 6.3VAC windings of the 150V source follower mains transformer, which will not have a time delay. The 30 second time delay in addition to the built-in soft start of the Maida regulators will give the filaments more than enough time to come up to voltage before B+ is applied.

You are a braver man than I am : D
 
Jun 14, 2020 at 3:24 PM Post #479 of 6,723
The eagle has landed, the Omega Super 8 High-Output Monitors have arrived. The veneer is figured walnut.

IMAG2165-2.jpg

They were waiting for me when I got home from work last night, of course I threw them in the stereo and gave them a quick listen despite being completely exhausted, more listening this morning and throughout the day. Omegas are notorious for their long break-in period (or brain-in?), so I will defer giving major impressions until later, maybe past the 100-hour mark. I will say this right off the bat - the midrange detail and vocal nuance is impressive. They are also much more sensitive to on-axis positioning than my Snells. From my short positional experiments, I've found they need a pretty generous toe-in. Their sound is also more sensitive to changing tube configurations than the Snells.

IMAG2164-2.jpg

From a DIY standpoint, I can now do Snell surgery. Will have to decide where this project falls in terms of priority with the 841/801A amplifier prototype underway, but I plan to re-cap and re-resistor the crossovers and replace the tweeter ferrofluid.

Beautiful wood.I would have thought it was maple though.
 
Jun 14, 2020 at 4:46 PM Post #480 of 6,723
801A B+ supply is done now, adjustable to 450V on the Maida reg. Change of plans on the 801A mains transformer - the toroid was just too big, swapped it out for an LL1650.

IMAG2213.jpg

Put together one of the time delay boards as well, it works perfectly. After ~25 seconds, the relay closes and the 120VAC hits the mains transformer. The Maida reg then ramps up its voltage over another 10-15 seconds to ~450, more than enough time for the 841 / 801A filaments to heat. The 801A supply will drop another 25-30V across the output transformer primary, giving me my 420V 40mA bias point.

IMAG2211.jpg

Just have to put the other time delay kit in place and the power supply is done. I love it when everything goes according to plan :)

With the B+ supplies ready to go, next step will be putting together the source follower circuit and bias wiring on one of the proto-boards. Have to work the next few days, so it will be a project for next weekend most likely. This is coming together much more quickly than I expected, still another 2-3 weeks out from getting the output transformers, good thing I have other projects to work on.
 
Last edited:

Users who are viewing this thread

Back
Top