How does one even purchase a Transcendar these days? Their web site seems to be devoid of pricing information and an obvious way to buy. Seems like the selection of available configurations has also decreased significantly.
A modified Magnavox 8802 in the Gillespie style
How does one even purchase a Transcendar these days? Their web site seems to be devoid of pricing information and an obvious way to buy. Seems like the selection of available configurations has also decreased significantly.
Lavane
Lunatic Member
I noticed this as well. It has me scratching my head. They have the ones I want but like you stated how do you buy them? I guess contact them? The last time I bought some I had to contact them to get a price including shipping. Then send a check.
Lavane
Lunatic Member
I just sent an email off to Transendar. The only contact info I could find on the whole site was on the customer amplifier page. I asked why there was no price info anywhere on their site and why the only contact info was on the costomer amp build page. I like their iron and for me shipping is fast as I live between 2 to 3 hours away from them. But if you have to dig theough every page just to find an email address to ask prices and availability im not sure I want to have to go through all that just to buy something. Edcor is looking good to me right now.
@Lavane, I see you're in "northern cal.", so I guess that means they're not terribly far away from me, either. They *used* to have prices for stuff on the web site, so this is a recent change. Wondering if they're having trouble staying afloat?
In any case, I am a big fan of Edcor ... the only think I don't like is the lead time!
In any case, I am a big fan of Edcor ... the only think I don't like is the lead time!
Lavane
Lunatic Member
The lead time is why I tried Transendar to begin with. You may be right. They sell on the bay as well. Might be worth a look to see Iif there is a fire sale going on.
Bummer! I wanted to give their products a try........
Dave
Dave
I know this thread is about a single ended amp, but since we are on the subject of Transcendar output transformers: I recently tried their least expensive push pull output transformers (for an EL84 amp) and thought their frequency response was okay, but I ended up sending them back because with the amount of feedback I wanted to apply, I had a dickens of a time getting the amp stable.
I did ask the owner in an email recently if his Fisher replacement transformers executed the exact same winding methodology, wire size, core size, core material, etc, as the original old stock iron for which the replacement transformer is replacing. The answer lies in the term "replacement" and not "clone." So in short, the answer was no.
If true Fisher clone transformers could be made, I think there would be a steady stream of customers knocking at the door...
I did ask the owner in an email recently if his Fisher replacement transformers executed the exact same winding methodology, wire size, core size, core material, etc, as the original old stock iron for which the replacement transformer is replacing. The answer lies in the term "replacement" and not "clone." So in short, the answer was no.
If true Fisher clone transformers could be made, I think there would be a steady stream of customers knocking at the door...
Push-pull!
. And you would think I would have remembered since I previously commented on your phase splitter approach. 
Well, the transformers arrived today. I'll have to get some fender washers or something to make a couple of the mounting locations work (due to larger holes already in the chassis), but they fit pretty well. Not sure I'll have any time to work on it this week / weekend, but light is at the end of the tunnel 
Attachments
Ok, to I managed to scare up a couple of hours to tinker with it last night, so I installed the output transformers and fired it up. Initial results are pretty good... with a single channel driven, it can do 12W at 1KHz before clipping. I had estimated the open loop gain at roughly 39, and it measured at 33.7, and the 1.2K feedback resistor thus works out on paper to -11.1dB of feedback. 1KHz square waves had a pretty big overshoot, to I added a 330pF feedback cap to tamp them down a little as a start. I'm definitely going to need to put some time into the HF tuning, because 8KHz square waves have a pronounced ring that keeps going for most of the wave top (1KHz settle down relatively quickly, but still overshoot a fair amount).
I don't currently have the bucking configuration on the PT ... the plate supply voltage is about 10V higher than I expected, but the screen and AF amp supply voltages are just about spot on. The PT does get a bit warm, but not really any warmer than the PT on my mostly-stock 8802 that uses the standard power supply configuration. There is a noticeable mechanical hum from the PT (though not especially loud), so I may revisit the bucking configuration if there's a chance that'll tamp that down. I have a digital volt/ammeter between my variac and the outlets I plug devices under test into, and at 120VAC it draws ~640mA at idle.
The cathodes are currently biased at +22V rather than my calculated +19V, and the current across the 10 ohm cathode resistor is 115mA, for an average of 28.75mA per tube. I haven't spent any time dialing that in yet. When I get a chance I'll measure the plate current of each tube individually (across it's OT winding) and probably shuffle them around to balance out the pairs as best I can.
The tubes that are installed are the original 6V6s that I received with the amp, and NOS Russian 6Н2П for the AF amp / PI. (One of the 6EU7s that came with the amp was bad, so I just went ahead and reconfigured the sockets.)
No distortion or frequency response measurements yet... I was already up a little later than I wanted to be, so that'll have to wait for another day.
But I did listen to some Beatles and Seal on it while cleaning up the bench, and it sounds pretty good, and with no signal, I could hear no hum or noise coming from the (fairly efficient) speakers it was connected to.
Updated schematic attached. (Well, ok, maybe not... file uploads don't seem to be working for me right now...? I'll try again later...)
I don't currently have the bucking configuration on the PT ... the plate supply voltage is about 10V higher than I expected, but the screen and AF amp supply voltages are just about spot on. The PT does get a bit warm, but not really any warmer than the PT on my mostly-stock 8802 that uses the standard power supply configuration. There is a noticeable mechanical hum from the PT (though not especially loud), so I may revisit the bucking configuration if there's a chance that'll tamp that down. I have a digital volt/ammeter between my variac and the outlets I plug devices under test into, and at 120VAC it draws ~640mA at idle.
The cathodes are currently biased at +22V rather than my calculated +19V, and the current across the 10 ohm cathode resistor is 115mA, for an average of 28.75mA per tube. I haven't spent any time dialing that in yet. When I get a chance I'll measure the plate current of each tube individually (across it's OT winding) and probably shuffle them around to balance out the pairs as best I can.
The tubes that are installed are the original 6V6s that I received with the amp, and NOS Russian 6Н2П for the AF amp / PI. (One of the 6EU7s that came with the amp was bad, so I just went ahead and reconfigured the sockets.)
No distortion or frequency response measurements yet... I was already up a little later than I wanted to be, so that'll have to wait for another day.
But I did listen to some Beatles and Seal on it while cleaning up the bench, and it sounds pretty good, and with no signal, I could hear no hum or noise coming from the (fairly efficient) speakers it was connected to.
Updated schematic attached. (Well, ok, maybe not... file uploads don't seem to be working for me right now...? I'll try again later...)
Managed to upload it now... seems it was complaining about the file name, or something. Anyway, this is "near final".
@RUmad -- heater voltage without the bucking configuration is 6.7V with my variac holding steady at 120V. On the high side, but within tolerance.
Took some frequency response measurements... using 1KHz @ 1Wrms as a reference, it's completely flat from about 40Hz - 20KHz, down 1dB at 30Hz, and down 2dB at 20Hz. This is about what I expected from these transformers. What I did not expect from these transformers is what happened after 20KHz... output kept increasing until I hit about 87KHz, at which point it was +2dB, after which it started falling off again. The two transformers also do not perform equally... one channel has pretty good looking square waves even at 8KHz and 10KHz with the feedback loop as drawn, whereas the other keeps ringing almost throughout the pulse. That same channel also gets wonky at the bottom of the sine wave right around clipping (sorry, no photos tonight... phone was out of juice).
The HF response characteristic is one of the weaknesses of modern OPTs. As with so many things, they just don't make them the way they used to! It would be interesting to know how much interleaving Edcor used in winding the coil form.
A suggestion for you: Try increasing the value of your 1K input grid stopper resistor into a range of about 47K to 100K. Because your level control is of such low resistance, the impact of its setting won't have a significant effect on the grid impedance, and the impedance of your source will aid in this even more. But increasing the grid stopper to as much as 100K will allow Miller to go to work just a little bit, which in this case is a good thing. By introducing more Miller into the input stage, it will effective reduce the response of the open loop circuit such that with the loop closed, you should be able to safely tailor the response of the final design below the resonant frequency of the output transformer, and get a handle on the excessive ringing. The resistor can be tweaked to suit.
Good luck with it!
Dave
A suggestion for you: Try increasing the value of your 1K input grid stopper resistor into a range of about 47K to 100K. Because your level control is of such low resistance, the impact of its setting won't have a significant effect on the grid impedance, and the impedance of your source will aid in this even more. But increasing the grid stopper to as much as 100K will allow Miller to go to work just a little bit, which in this case is a good thing. By introducing more Miller into the input stage, it will effective reduce the response of the open loop circuit such that with the loop closed, you should be able to safely tailor the response of the final design below the resonant frequency of the output transformer, and get a handle on the excessive ringing. The resistor can be tweaked to suit.
Good luck with it!
Dave
RUmad
Super Member
So, Mr Thorpe, to be clear; you tied all four cathodes together? That's how you come up with 115ma across the 10R resistor? I actually followed your schematic, as far as the EFB goes, but used two separate 10R resistors. I meant to ask you, and get Dave's approval, before I got started but I ended up diving right in this morning. Before I fire it up I just wanted to check with you guys, make sure I get the bias adjusted properly. Thanks!
Yes, I tied all 4 6V6 cathodes together. Basically, shoot for ~30mA of cathode current per 6V6. It might be a little more or less. Sorry, the schematic says "30mA plate current", but it should say "cathode current" because it obviously also includes what the screen is drawing, too.