• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

One Electron vs. Transcendar

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm busy collecting iron for a series of tube amp proquects, and am looking at a pair of 3k output transformers for use with a SE amp using a triode connected 6550. I've narrowed things down to the One Electron 3k transformer and the Transcendar "300B" 3k. Any comments pro or con?
 
Edcors have only one set of taps, which will make you choose a Secondary impedance to stick with. and also a primary impedance, Transcendar lets you pick different impedance taps. It just seems more practical to me. I would have orderd a pair already, were i not moving in two weeks.

-Alex
 
Hi, checking in for the first time an a week, on a borrowed slow dial up in rural West Virginia. Im leaving for the trip south tomorrow, so I may not be on for a few more days unless I find a hotel with free internet.

You probably know my story on this, but I will repeat it. I ran a listening test with these two transformers about a year ago. With 5 testers, all of whom owned tube amps, no one could reliably tell these two transformers apart. Some listeners could tell them from the Electra-Prints, but it was not a fair test since the Electra-Prints were 5K and the others were 3K. I liked the Transcendars enough to buy 10 pairs of them. Most of them reside in customers 300B amps now. I have had them in my SimpleSE, and they do sound nice, but the Hammond 1628SE's have far more bass. The Hammonds are twice as big as the other two. I will be testing the new 1628SEA's when I get back.

The Transcendars have 4 and 8 ohm taps, while the One Electrons have 4, 8, and 16 ohm taps if that makes a difference.

There is a page on my web site outlining these tests, but I can't find it now. This computer will only open one window at a time.
 
Tubelab, your test results have been pointed out - thanks. I was looking for some more data points, though I'll probably get the Transcendars on grounds of cost. I consider ordering the Hammond 1628s to be a cra* shoot at this point until the old transformers are flushed out of the system.

The "expensive" Edcor SE transformers are interesting in that they are offered down to a really low primary impedance. This may make it possible to do a monster SE enhanced triode sweep tube amp using a fairly low B+ voltage. That's an experiment for the future, as I have more projects than I can handle as it is:

1) 6CD6GA SE enhanced triode - in progress using Hammond 125ESE outputs. Awaiting final circuit check and first AC power-up.

2) 6GF7A SE with switchmode supply - in progress using Hammond 125CSE outputs. The switcher is done - next step is to start drilling all the holes for the tube sockets, transformers, terminal strips, etc.

3) 6GF7A push-pull (will configure chassis on the 6GF7A SE project so I can try this fairly quickly) I have cheap Edcor 8K 10W XFMRs for this one.

4) 7591A P-P using old Fisher iron and EH output tubes.

5) 6HV5A SE - I'm leaning strongly toward using 700V B+, A2 mode and paralleled output tubes for this one, using the 5K Transcendar transformers I bought a while ago. I was thinking of using >1k B+ and 10k output transformers, but I decided I wanted to die of old age... Matching the 6HV5As should be interesting - I don't even know if my tester will handle them. Even in A2 mode, the drive requirements for the 6HV5As should be pretty low, so I am thinking of using a 6CM7 for preamp/follower drive.

6) 807/1625 SE Ultralinear - I have the cheap Edcor 5k iron for this one. This one may move up in priority if I chance across an inexpensive chassis of the right size.

7) 6550 triode strapped SE - this one will use the 3k output transformers mentioned above.
 
wrenchone, I wish I had time for that many experiments. I have been down these roads, or ones close to them, so I will relate my experiences.

7) 6550 triode strapped SE - this one will use the 3k output transformers mentioned above.

I have put 6550's in the SimpleSE and tried them in triode and UL mode with several transformers. Triode works well with the One Electron or the Transcendar (hard to tell them apart). I have used Sovteks and NOS GE's the GE's really rock but they are expensive.

6) 807/1625 SE Ultralinear - I have the cheap Edcor 5k iron for this one. This one may move up in priority if I chance across an inexpensive chassis of the right size.

I tried 6L6GB's in the same amp with the $18 Edcors. I preferred the ability of 6L6GC's to dissipate more power, and thus handle more current.

5) 6HV5A SE - I'm leaning strongly toward using 700V B+, A2 mode

I only have 2 of these. I wired them into a TubelabSE, but I only had 400 volts of B+. I couldn't get them to draw enough current to make any power. That amp couldn't do positive grid voltage.

4) 7591A P-P using old Fisher iron and EH output tubes.

I have an old Fisher receiver that uses 7591's. I played with it for a while, but wound up stealing the tubes for my old Guild guitar amp.

2) 6GF7A SE with switchmode supply

Tried 6EM7's and 6DN7's. Not enough power for my wimpy 86db speakers. maybe 1.5 watts. Tell us more about the SMPS though.

I just got back from a 3000 mile road trip. There is a pair of the new Hammond 1628SEA's waiting for my attention. Unfortunately when I got to work today, I found out that the test IC's that we sent out came back early (never happens) so I must spend the next month evaluating them. I will try to test the transformers this weekend.
 
Tubelab - I you eventually go out and try to mess with the 6HV5As, hold out for GE or Westinhouse. The plate structures on these two are much larger than any of the other brands I have on hand, including RCA, Sylvania, ECG, Radio Shack, and Zenith. I would expect them to be able to deliver more power in a pinch.
 
Tubelab, a few comments about your comments...

1) 807/1625 Ultralinear SE - I have 807s and 1625s on hand, so that's what I'll be using - not a 6L6 in the house. Depending on whose data sheet you read, the screen rating is fairly decent for these tubes - important, as I want to push the B+ as high as I can. I'll be starting at around 400V, though I expect some attrition from drops through the transformer winding and cathode bias resistor. I'm using the $18 Edcors, too - an experimenter's dream come true.

2) 7591A P-P using Fisher iron - When I'm done, the only similarities between the original Fisher circuit and mine will be the iron and the output tubes. I'm looking at using a 6CM7 with a folded cascode front end and concertina splitter or a more conventional circuit using the 5965. I used to have an old Scott LK-72B back in the early '70s, which used 7591 outputs for about 30W/channel. That amp souded pretty sweet to me, even using the old 6U8 driver circuit.

3) 6GF7 SE and P-P - Ths switching power supply I'm using takes a rectified low voltage input and boosts it to 340VDC for the B+, and a -75V bias supply. I had a choice between using a ZVS flyback operating directly off the mains or the DC-DC approach. I chose the DC-DC, as I could whomp something together fairly quiclky out of parts at hand, and I do enough experimental dinking around with line input switchers at work to make me reluctant to pusue the madness at home as well.

The business end is a choke-fed saturating transformer push-pull converter driven by a buck stage. The control loop is closed around the buck. The control circuit is a CD4093 CMOS Schmitt trigger and a handful of components. The overall power supply efficiency is 77-78%. This is not great, but I don't care too much, as I threw any notion of efficiency straight out of the window anyway once I chose a class A amplifier...

The filaments of the 6GF7 will be hooked in series and driven from the input DC supply with an LM317 current source - another efficiency killer. I'll have to rethink the scheme a bit when I throw another pair of tubes on to the chassis to try the push-pull design. I'll post a schematic of the switcher when I actually have it satisfactorily running a pair of 6GF7s. Right now, the only thing it's driven is a 5k resistive load.
 
I you eventually go out and try to mess with the 6HV5As, hold out for GE or Westinhouse.

The two that I have are NIB GE's I may have more buried in the warehouse but I don't know. I will get back to these sometime.

I am anxious waiting for your test results of Hammond's 1628SEA. I just wish this new Transformer will clear all the bad words about them for a long long time.

I ran a quick frequency response plot on one of them. The response gently rolls off starting at 12KHz. It is 2 db down at 20 KHz. All of the distortion at high frequency stuff is gone. I plan to do more testing this weekend.

807/1625 Ultralinear SE - I have 807s and 1625s on hand, so that's what I'll be using - not a 6L6

It has been debated back and forth several times but I believe, and I have taken several of each apart (already broken), that an 807/1625 IS a 6L6GB in a different envelope. I have several 6l6GB's and a lot of 807's and 1625's. For a given manufacturer and date the internals for these 3 tubes are identical and they test the same. The higher plate voltage ratings are due to the plate cap. These tubes can handle about 20 watts of dissipation while a 6L6GC can take 30 watts. I have already taken the 6L6GB's to 450 volts B+ (cathode bias, 410 across the tube) without issue. The limiting factor becomes dissipation from me getting carried away with the current. 6L6GC's allow me to turn it up more, and I can get them for $4 each (Ebay).

I used to have an old Scott LK-72B

I have a Scott 272 it uses EL-34's and that same phase splitter only it is a 6BL8. It does sound nice. I still have the Fisher (a 400 receiver I think) stashed in the warehouse and I may "borrow" its transformers some day but it works and is in nice condition, so I hate to mess it up.

6GF7 SE and P-P

So the whole amp runs on 12 volts? My attempts in the SMPS world have not been highly sucessful. I did a 12V to 28V boost converter (200 watt) at work that is going into production now. I tried to scale up the output voltage and then drive a voltage multiplier, but 150 to 200 volts seem to be the limit without a transformer. I tried to build an offline converter uning an ON semi chip, but my attempts turned out to be instant sacrifices to the fire gods! For now I'll stick to conventional transformers.
 
Re Tubelab Comments:

1) 1625/807 SE UL - I'm resisting the temptation to buy "just one more tube", and will try to do the best with what I have on hand. I have to draw the line somewhere. Anyway, if I wanted to gild the lily octal style, I'd just use the other pair of 6550WEs I have on hand.

2) 6GF7A amps with switching supply - the DC-DC converter I've designed runs off of regualted and filtered 12VAC, so the actual DC will be higher. I need to test with actual AC transfiormer input to adjust the DC-DC converter loop compensation for maximum 120 Hz ripple rejection. So far, all my testing has been with a bench DC supply and resistive loading. I had to wind my own transformer for this baby. I didn't think a multiplier was a viable approach, and the transformer took me all of about 1-1.5 hrs to make. My speakers are pretty efficient with 16 drivers on a side. It'll be interesting to see how much of a noise 1-2W can make - I've never really tried to measure my speaker sensitivity.
 
Transcendar regularly sells pairs of their 5k and 3k output transformers on Ebay. I can't remember the exact seller name, but if you search this site for other mentions of Transcendar, you'll find it. I just load Ebay with "output transformer", and I eventually find the seller if he has anything up for auction. The price is pretty good even if you use the "buy it now" option instead of duking it out at the auction close. I've never tried contacting Transcendar directly (even though they're close by), but that's obviously what tubelab did when he bought multiple units for his amplifiers. From what I have heard, the units offered are production overstock and when they're gone, they're gone. Correct me if I'm wrong on this one. Ebay and direct contact are the only two ways I've seen of getting these transformers - no distributors that I know of.
 
I just recently purchased a pair of 3K ohm SE Transcendars at auction on eBay. The seller's name is *ggaet* which is Gery Gaetani of Transcendar. (831-636-0404) They claim to have a new design that now supersedes the exhausted earlier stock. This new design extends the high end out to over 80KHz, where the earlier ones rolled off at around 38KHz.

Additionally, there are also 3.3K and 5K ohm versions of this new SE design with accompanying extended response over the old. The higher impedance units have more inductance with a slightly lower top end as would be expected. (normal design trade offs)

When my transformers arrived I noticed the cores were wrapped with a single layer of tightly woven black cloth around the outside. I'm assuming this is for cosmetic purposes as it does look nice, but hides the laminations from view. While I've never seen Transcendars before, I don't know if this is standard or something new.

Since I have a quantity of WE 300Bs I figured it was time to use them for something. Unfortunately it won't be for a while with my busy schedule.

Victor
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.