[L0rdGwyn]

Probably a stupid comment, so brace yourself.

Have you tried looking for the tweeters on ebay? I usually find decent deals on j3 tweeters. IDK about j2 tweeters.

I haven't looked much, but at a glance it seems like J/III or OEM equivalent parts are sold pretty often on eBay. I think the J/II tweeters are sort of hard to come by, especially in good shape since the doping eventually dries out. Atomic Hifi is sending me some tweeters, I doubt they will be original J/II though, mabye J/III, but better than having a mismatched pair.

 

[Tjj226 Angel]

I haven't looked much, but at a glance it seems like J/III or OEM equivalent parts are sold pretty often on eBay. I think the J/II tweeters are sort of hard to come by, especially in good shape since the doping eventually dries out. Atomic Hifi is sending me some tweeters, I doubt they will be original J/II though, mabye J/III, but better than having a mismatched pair.

Well just as a heads up, I have like 6 pair of j3 tweeters from ebay. None of them are "matched". They all sounded different out of the box, but once I went through and refurbished them, they were all well within tolerance of one and other.

A friend of mine has the J2s. He refurbished his tweeters and all the little different imbalances and nuances to the sound were cleaned right up.

 

[L0rdGwyn]

Well just as a heads up, I have like 6 pair of j3 tweeters from ebay. None of them are "matched". They all sounded different out of the box, but once I went through and refurbished them, they were all well within tolerance of one and other.

A friend of mine has the J2s. He refurbished his tweeters and all the little different imbalances and nuances to the sound were cleaned right up.

Thanks, how did you refurbish besides replacing the ferrofluid? I had mentally committed to moving on, but I might keep an eye out for a match to one of my two tweeters, would be preferable to use the originals. From what I have gathered, one is an earlier model made by Culver that was used in the E original (cannot confirm if/when it was used in the J or J/II), Snell later switched to Tonegan when the Culver part became obsolete, which appears to be the make of my second tweeter that was used in both the J and E. They are physically identical as far as I can see, I do have to wonder if they are equivalent parts and someone just didn't add ferrofluid to one mistakenly, hence the difference in HF response...coming up short on Google though.

Culver S-90662



Tonegan 94C70DS-00HD

 

[Tjj226 Angel]

Thanks, how did you refurbish besides replacing the ferrofluid? I had mentally committed to moving on, but I might keep an eye out for a match to one of my two tweeters, would be preferable to use the originals. From what I have gathered, one is an earlier model made by Culver that was used in the E original (cannot confirm if/when it was used in the J or J/II), Snell later switched to Tonegan when the Culver part became obsolete, which appears to be the make of my second tweeter that was used in both the J and E. They are physically identical as far as I can see, I do have to wonder if they are equivalent parts and someone just didn't add ferrofluid to one mistakenly, hence the difference in HF response...coming up short on Google though.

Culver S-90662



Tonegan 94C70DS-00HD

Ooooooookkkkkaayyyyy maybe someone at the snell factory knew something we dont?

Those two tweteers do indeed look identical. Maybe the deal is that tonegen bought culver? Maybe they are physically the same part but just with different company labels on them?

Its super unlikely that two tweeters from two different companies would look nearly identical. There is more to this story than what we know.

I would take a business card and soak it in some alcohol. Then insert it into the tweeter that doesn't have fluid and try cleaning out any crap in the cavity. If the card comes up with this redish brown color, then the tweeter had fluid in it at one point. I have a real funny feeling both of these tweeters take fluid, and they are identical tweeters.

I would clean out each tweeter really well anyways. I always end up finding a surprising amount of dirt in the drivers.

 

[L0rdGwyn]

Ooooooookkkkkaayyyyy maybe someone at the snell factory knew something we dont?

Those two tweteers do indeed look identical. Maybe the deal is that tonegen bought culver? Maybe they are physically the same part but just with different company labels on them?

Its super unlikely that two tweeters from two different companies would look nearly identical. There is more to this story than what we know.

I would take a business card and soak it in some alcohol. Then insert it into the tweeter that doesn't have fluid and try cleaning out any crap in the cavity. If the card comes up with this redish brown color, then the tweeter had fluid in it at one point. I have a real funny feeling both of these tweeters take fluid, and they are identical tweeters.

I would clean out each tweeter really well anyways. I always end up finding a surprising amount of dirt in the drivers.

Yeah my thoughts exactly, pretty bizarre. I'll see if I can dig up other information, but I'm going to clean them up, put ferrofluid in the one without and give them a try, I'll have the other tweeters as backups if things don't sound right.

 

[L0rdGwyn]

Got the input stage operational with filament bias. Had to make some changes to account for the additional voltage dropped, changed transformers and some components on the Coleman regs.

What we ended up with is a regulated 650V B+ at the top of the CCS, 484V plate-to-cathode, a 9mA plate current and 5.15V on the filament with 4ohms resistance.



If it looks like a rat's nest, that's because it is. I am still using alligator clips predominantly as it gives me the flexibility to choose which power supplies are connected at any one time. Once everything is verified and semi-finalized, I'll solder it all together.

Tomorrow I am going to throw some sine waves into the 841 stage and source follower, we'll see what comes out the other side. Theoretically, I should have the output transformers in a little over a week, that's when things will really get really interesting. With the changes made to the filament supply, I now have an entire set of spare parts for another pair of 10Y / 801A tubes: DC supplies, Coleman regs, and filament transformers (whoops!). Those are going to burn a large hole in my pocket, we'll have to come up with a plan for them

 

[Tjj226 Angel]

Got the input stage operational with filament bias. Had to make some changes to account for the additional voltage dropped, changed transformers and some components on the Coleman regs.

What we ended up with is a regulated 650V B+ at the top of the CCS, 484V plate-to-cathode, a 9mA plate current and 5.15V on the filament with 4ohms resistance.



If it looks like a rat's nest, that's because it is. I am still using alligator clips predominantly as it gives me the flexibility to choose which power supplies are connected at any one time. Once everything is verified and semi-finalized, I'll solder it all together.

Tomorrow I am going to throw some sine waves into the 841 stage and source follower, we'll see what comes out the other side. Theoretically, I should have the output transformers in a little over a week, that's when things will really get really interesting. With the changes made to the filament supply, I now have an entire set of spare parts for another pair of 10Y / 801A tubes: DC supplies, Coleman regs, and filament transformers (whoops!). Those are going to burn a large hole in my pocket, we'll have to come up with a plan for them

Stop buying 801a tubes damn it! I need them to come down in price : P

But in all seriousness, im surprised how well this is coming together. When we were discussing this originally, I knew this design would be a huge pain in the butt, but you seem to enjoy crazy projects, so I will have to keep that in mind for our next talk lolololol.

I can just see you trying to make an OTL speaker amp as your next project.

 

[L0rdGwyn]

Stop buying 801a tubes damn it! I need them to come down in price : P

But in all seriousness, im surprised how well this is coming together. When we were discussing this originally, I knew this design would be a huge pain in the butt, but you seem to enjoy crazy projects, so I will have to keep that in mind for our next talk lolololol.

I can just see you trying to make an OTL speaker amp as your next project.

Yeah, this amp is a little nuts lol but it has all gone according to plan thankfully, fewer pitfalls than I expected. I spent a lot of time running sims, researching, troubleshooting, so far it has paid off. I experimented with every possible iteration of single-ended 801A I could find, didn't know at the beginning what a PITA tube it is to use, the 841 too, we'll see if I made the right design choices when the OPT arrive.

I don't want to build an OTL speaker amp LOL too big, too hot, too crazy. Not sure what will be next, I've considered working on something cost-effective and not insane that I can build for other Head-Fiers, I think that might be fun, I am reaching amplifier saturation and SAF (spouse acceptance factor) is getting a little dicey. I might have a DIY speaker build in the near future but that is TBD.

 

[whirlwind]

SAF.....LOL!

Shoot me a pm, I may have a small project for you at some point

 

[L0rdGwyn]

Wired up for some measurements today, specifically want to know the gain of the input stage. I try to keep a DMM across a HV portion of the circuit at all times so I don't kill myself, yeah that's 708V across the voltage doubler resevoir cap takes a minute or two for the caps to discharge across the bleeder.



With a 5.6Vpp input, I am getting a 164Vpp output. This represents the maximum input signal from a typical DAC (5.6Vpp = 2Vrms, standard single-ended DAC output voltage). I also figured out how to take screenshots on my scope instead of taking pictures of the screen like a caveman.



That puts the μ of the 841 at a little over 29 in this configuration. Nice. Will wire up the source follower next.

 

[L0rdGwyn]

Wired up the source follower and did the same, 5.6Vpp input. The output of this circuit will be direct-coupled to the grid of the 801A, so will provide the current drive into A2 region as well as the fixed bias.



As you can see, I have my approximate -25V bias for the 801A grid, this can be adjusted with one of the two trimmers on the board (this will be replaced with a full-sized potentiometer accessible from outside the chassis for bias adjustment in the final build).

Here is the output, essentially identical to the input (plot above), 164Vpp into the grid of the 801A. This is beyond what is needed for max output swing as the 801A should hard clip at plate saturation, around +45V on the grid, roughly 130Vpp.



Everything works. At this point, all there is to do is wire things up and wait for the OPT to arrive to set up the output stage, this prototype is nearing completion.

 

[L0rdGwyn]

I mentioned in a previous post that I was thinking about working on a (relatively) inexpensive headphone amplifier design that I could build for other Head-Fiers.

Well I sort of ran with the idea and today came up with the skeleton of a design. The tube I penciled in for this idea way back was the 1626. The datasheet doesn't show it very well, but this tube is very linear and a good choice for a low power SET (like a headphone amp), it is widely available, used in the Ampsandsound Kenzie as well as the Toolshed Darling headphone amps. My idea is a little different.



With the OPT I am considering, a relatively small amount of gain would be needed to drive headphone to ear-destroying volumes. The preliminary idea is a 1626 driving 1626 parafeed headphone amp with taps for 32, 150, and 300ohms. Matched quads of these tubes are very easy to find. With the tube driving itself, the distortion spectra will cancel to some degree, with a parafeed output should be a very clean sound.

I am getting ahead of myself, again, but I think I am going to prototype one of these, will be a cakewalk compared to what I have been working on! If it sounds good, I might consider building one for real and sending it on a tour.

 

[Zachik]

If it sounds good, I might consider building one for real and sending it on a tour.

So, how do I sign up for the tour?

 

[L0rdGwyn]

So, how do I sign up for the tour?

If it comes to fruition, you just did!

 

[audiofest2018]

I mentioned in a previous post that I was thinking about working on a (relatively) inexpensive headphone amplifier design that I could build for other Head-Fiers.

Well I sort of ran with the idea and today came up with the skeleton of a design. The tube I penciled in for this idea way back was the 1626. The datasheet doesn't show it very well, but this tube is very linear and a good choice for a low power SET (like a headphone amp), it is widely available, used in the Ampsandsound Kenzie as well as the Toolshed Darling headphone amps. My idea is a little different.



With the OPT I am considering, a relatively small amount of gain would be needed to drive headphone to ear-destroying volumes. The preliminary idea is a 1626 driving 1626 parafeed headphone amp with taps for 32, 150, and 300ohms. Matched quads of these tubes are very easy to find. With the tube driving itself, the distortion spectra will cancel to some degree, with a parafeed output should be a very clean sound.

I am getting ahead of myself, again, but I think I am going to prototype one of these, will be a cakewalk compared to what I have been working on! If it sounds good, I might consider building one for real and sending it on a tour.

I am signing up for its tour as well!