Some details for you to consider.
From George at tubelab, He suggested that R18 and R19 be 2W, based on his experience of blowing them up. Also R5 and R6 should be at a minimum 3W, if not 5W for margin.
You have a PM
From George at tubelab, He suggested that R18 and R19 be 2W, based on his experience of blowing them up. Also R5 and R6 should be at a minimum 3W, if not 5W for margin.
You have a PM
Slight diversion in thoughts here. From my tests, to have sufficient 'headroom' for the ECC88 and to place the maximum signal swing in the middle of that headroom, the cathodes of the ECC88 cannot be much higher than some 82Vdc. This is what must be sought out for those bent on a triode input stage (can they say why?). A triode input will also lower the loop gain - but not necessary into unpractical realms. We will see.
Eli and I are at present exchanging PMs; he is in favour of buffering the ECC88 output with FET-followers, with obvious advantage; it depends on how much of a mix one wants a tube amp to be. One other slight difference: The effect of the input stage Miller capacitance is not ruled by R1, it acts on the output impedance of the feeding pre-amp/controller/volume-control/whatever, which is hopefully much lower. One merely uses a high value so as to not run into any danger of loading that 'feeder's' impedance, and as the grid must be tied to common by something.
[But I hope not just a volume control; my previously stated problem with a triode here is that the mentioned Miller-C acting on a variable input resistance can influence the NFB stability. Even if the resulting pole is above 20kHz - it usually is not by much - the effect of a variable Zin on a square wave can easily be demonstarted. That to me is the main advantage of using a pentode input.]
Oh, lastly. There is no necessity to use an LED also in the first stage cathode circuit. If the dc bias value obtained is too low (as it might be for a triode) one simply raises the value of the cathode and NFB resistors. For all its advantages, I am worried about the linearity of an LED there (to be honest I should have said I wonder; I have no info either way.) Point (not often realised): Any non-linearity in the process of mixing input and NFB signals, is NOT cancelled by said NFB. [Not to be 'wise', but his is often the downfall in semiconductor amplifiers with their often quite high factors of gNFB. "Feedback is bad"..... no, it is simply the designer falling short. Sorry for being blunt.]
Eli and I are at present exchanging PMs; he is in favour of buffering the ECC88 output with FET-followers, with obvious advantage; it depends on how much of a mix one wants a tube amp to be. One other slight difference: The effect of the input stage Miller capacitance is not ruled by R1, it acts on the output impedance of the feeding pre-amp/controller/volume-control/whatever, which is hopefully much lower. One merely uses a high value so as to not run into any danger of loading that 'feeder's' impedance, and as the grid must be tied to common by something.
[But I hope not just a volume control; my previously stated problem with a triode here is that the mentioned Miller-C acting on a variable input resistance can influence the NFB stability. Even if the resulting pole is above 20kHz - it usually is not by much - the effect of a variable Zin on a square wave can easily be demonstarted. That to me is the main advantage of using a pentode input.]
Oh, lastly. There is no necessity to use an LED also in the first stage cathode circuit. If the dc bias value obtained is too low (as it might be for a triode) one simply raises the value of the cathode and NFB resistors. For all its advantages, I am worried about the linearity of an LED there (to be honest I should have said I wonder; I have no info either way.) Point (not often realised): Any non-linearity in the process of mixing input and NFB signals, is NOT cancelled by said NFB. [Not to be 'wise', but his is often the downfall in semiconductor amplifiers with their often quite high factors of gNFB. "Feedback is bad"..... no, it is simply the designer falling short. Sorry for being blunt.]
Last edited:
Johan
I just wanted to see if the switch to the ECC88 opens up any doors for triode use. If the only viable option is the EF86 in pentode than that shall be the end of the story. BTW, a CCS on the triode anode is acceptable to me if the results would prove better than the EF86 in pentode mode.
Let me know your final analysis, and conclusion, and I will finalize this design stage! I'm ready to move on.🙂
Also, I rather not use FET buffers, and would like the output stage to stay completed as is. The added complexity for this beginer project would be too great. SGregory has already completed a nice design for those exploring that path, and doesn't mind the additional work.
Johan, thanks again for your help in wraping this all up. That also goes for Eli , and everyone else too.
I just wanted to see if the switch to the ECC88 opens up any doors for triode use. If the only viable option is the EF86 in pentode than that shall be the end of the story. BTW, a CCS on the triode anode is acceptable to me if the results would prove better than the EF86 in pentode mode.
Let me know your final analysis, and conclusion, and I will finalize this design stage! I'm ready to move on.🙂
Also, I rather not use FET buffers, and would like the output stage to stay completed as is. The added complexity for this beginer project would be too great. SGregory has already completed a nice design for those exploring that path, and doesn't mind the additional work.
Johan, thanks again for your help in wraping this all up. That also goes for Eli , and everyone else too.
Last edited:
Being determined to lock down the main circuit, and PS by the end of this week, I have decided that transfomer purchases must begin due to the delay in them being constructed. I had decided on Electraprint a couple of weeks ago, after getting pricing from Jack, but had been wavering on the less expensive, but possibly better performing single secondary output option he offered. Seeing that SGregory had purchased the same exact Transformer, pushed me into that direction. So, they were ordered today, and the parts buying begins.
I plan to call Edcor on monday and order 2 XPWR002 Power transformers. The the price of the ElectraPrint option (that Sgregory quoted me), was more than I was willing to spend on the PT's. I'll have to make the Edcors work, and looks like i'll need to paint them black myself to match the Electraprints. Would I be bettter off requesting unpainted bells?
When the Transformers arrive, I can finally design a chassis. I want proffessoinal looking results. I'm still undecided on modern, or classic styling. One option would be the classic wood chassis / aluminum top plate. Going that route, I would build, or have built a wood box of walnut, and have Front Panel Express do the plates, as they have done excellent work on other things for me.
Open to all options on chassis construction, and parts selection as the builds begins.
I plan to call Edcor on monday and order 2 XPWR002 Power transformers. The the price of the ElectraPrint option (that Sgregory quoted me), was more than I was willing to spend on the PT's. I'll have to make the Edcors work, and looks like i'll need to paint them black myself to match the Electraprints. Would I be bettter off requesting unpainted bells?
When the Transformers arrive, I can finally design a chassis. I want proffessoinal looking results. I'm still undecided on modern, or classic styling. One option would be the classic wood chassis / aluminum top plate. Going that route, I would build, or have built a wood box of walnut, and have Front Panel Express do the plates, as they have done excellent work on other things for me.
Open to all options on chassis construction, and parts selection as the builds begins.
Last edited:
I think with Edcor you can get any color you want as long as it's blue.....I recently asked about black and the answer was no. I didn't ask about unpainted but I can't imagine that they would ship them unpainted as they will rust quite quickly.
OK, Time for some sandpaper.🙂
Boywonder, I may require you services one more time. The power supply as it now stands is in Tay's schematic revision 7. Looks like Johan is now recommending 288V to the EF86. If that tube is retained the numbers will need to be run again to drop voltage at the last stage in the PS. Hold off until we here from Johan again.
Also, reading back through this long thread. Some concerns about the Edcors voltage were voiced, but as best as I can determine, your last word was that you thought it should still be ok for about 465V. Is that right? Thread was getting confusing at that point.
A fellow had his blue Edcor end bells abrasively blasted and powder coated matte black. The result was superb, but I wonder about the expense.
Depends on the guy doing it.
The sand blasting is no problem to do yourself if you can get your hands on the gear. If you dont, I think the parts are small enough to do it by hand. And when I say 'by hand' of course I mean by machine 😀
Grinding down the laquering is propably easiest with a fan grinder or wire brush tool for a drill stand or dremel. From my gunsmith I learned that fine grinding goes fastest (and best if you're not increadibly skilled) with a abrasive-fleece-for-drills-thingy (like this: http://www.wuerth.de/de/presse/fpm09/bilder/schleifvliesfaltwalze-kl.jpg).
If you have the tools, youre propably can do one bell in ten minutes or so.
Matte laquering you can do yourself too. Gloss is where it gets problematic.
Powder coating is just a special way of laquering, only difference is that the particles aren't suspended in a solvent. It comes off just like everything else 😀
I believe Boywonder previously suggested just roughening up the surface instead of removing all the coating?
I painted an Edcor, just get a good uniform sand with about 220 then 400 grit. Test your choosen paint on a small corner to make sure it doesn't lift and you should be okay.
I attached a possible layout for the amp based on what I have read on what is important for a good mullard. As symetrical as possible. The input tube is not hooked up at this point as it hasn't been choosen. If this is what tubemack wants I will be happy to update based on the discussion here.
I attached a possible layout for the amp based on what I have read on what is important for a good mullard. As symetrical as possible. The input tube is not hooked up at this point as it hasn't been choosen. If this is what tubemack wants I will be happy to update based on the discussion here.
Last edited:
Regarding the 288V, I simply kept R1 and R2 what they always were, viz. R1=220R and R2=68K. The original voltages there were misquoted.
I'm not sure I understand? . Are you saying the PS will produce 288v as is? Are the other voltages correct?
I started with rusty grey end bells lifted from some rusty (left out in the Florida rain for a year rusty) HP power transformers that I wouldn't trust, even though they worked. I sanded all of the rust off (down to bare metal) and scuffed the non rusty areas with a Scotch Brite pad. Then I washed the end bells with lacquer thinner. Next I painted the bells with flat black Rust Oleum spray paint. It must be a paint capable of being used on bare metal (a primer). The flat black was left out in the winter Florida sun for an hour to dry. I let them cool and then followed with a thin coat of gloss black Krylon spray paint which was also sun dried with several additional coats applied to the warm end bells at 15 minute intervals. They were then sun dried for a few hours. The temperature was about 65 degrees during this process. This does not work on a typical 85 or 90 degree day.
While the end bells were drying, I took a pair of open frame guitar amp OPT's, taped off the brown paper windings, scuffed the varnish with Scotch Brite, and sprayed the laminations flat black.
The "granite" surface of the amp is actually aluminum sprayed with "stone" spray paint. The entire chassis for this amp was done with material from Home Depot.
Attachments
Hey TubeMack, I was researching the name Adagio (in order to create a nice logo design), and I have some bad news. There is already an Adagio tube amplifier, and it's a commercial one at that, so you might want to reconsider the name.
Adagio SET Tube Amplifier
With their prices, they wouldn't need to sell too many amplifiers to afford a powerful trademark litigation team.
..Todd
Adagio SET Tube Amplifier
With their prices, they wouldn't need to sell too many amplifiers to afford a powerful trademark litigation team.
..Todd
