I'm planning a 6GV8 headphone amp using current source loading on the triode stage, and wondering about an appropriate output transformer. Rails will likely be around 200VDC. I'll probably be driving Something similar to an Audio Technica M60-XA. I'll likely be shooting for 10X gain, closing the loop around the XFMR.
Are you planning parafeed or series? For parafeed there are the Edcor PCW's or Cinemag equivalents. For series feed, I believe the only options are from Lundahl or Sowter.
Maybe the Edcor XSE10-8K-50 ? That's got an 8k primary (UL tap included), 10W rated, 50 ohm secondary.
Its primary impedance isn't very high, but maybe if you enclose it in the gNFB loop it will give you enough low end response.
It's not expensive, at about $40 USD each.
https://edcorusa.com/products/xse10-8k-10w-8k-ohms-single-ended-tube-output-transformer
That's the only SE OPT makes with a 50 ohm secondary.
You could use an OPT w/ 16 ohm secondary, which you probably knew already, but I figured I'd throw that out there.
Maybe a 5k:16 OPT would work? https://edcorusa.com/products/gxse10-5k-10w-5k-ohms-single-ended-guitar-tube-output-transformer-1
Its primary impedance isn't very high, but maybe if you enclose it in the gNFB loop it will give you enough low end response.
It's not expensive, at about $40 USD each.
https://edcorusa.com/products/xse10-8k-10w-8k-ohms-single-ended-tube-output-transformer
That's the only SE OPT makes with a 50 ohm secondary.
You could use an OPT w/ 16 ohm secondary, which you probably knew already, but I figured I'd throw that out there.
Maybe a 5k:16 OPT would work? https://edcorusa.com/products/gxse10-5k-10w-5k-ohms-single-ended-guitar-tube-output-transformer-1
If I do a parafeed setup, I would prefer using a triode output, as I intend to load the output stage using a depletion mode mosfet current source. So, It would look a lot like the headphone Amp I designed using the 13DR7, a tube of which I still have ample stock. First stage and second stage would both employ depletion mode current source loading, likely with an LND150 on the first stage and a little IXYS/Littelfuse 500V TO-220 depletion mode part at the output.
In the interest of learning, is there a reason you prefer a triode in parafeed? Or in other words, is there a reason a pentode would be undesirable in parafeed?
Well - I dislike the idea of a pentode butting heads with a depletion mode mosfet current source, though the particular pentode in question does have a lowish plate resistance of 25k. I may prefer to use a pair of the (very pricey) Thermionics Lab headphone output transformers, which are kitted out with a 32 ohm tap. I would jigger with the screen voltage and grid bias to get a plate current in line with optimum bass performance.
I wasn't aware of Thermionic Labs. Their offerings for headphone amp transformers are very intriguing. No need for parafeed and several different primary options. They are indeed pricey though. I don't know if I want to spend that much on transformers for a headphone amp.
One thing that's been on my mind regarding the Edcor's, is the max input voltage. They list 15vrms for the PCW series. From what I've been simulating in Spice, it seems I can only get 50-100mW into 32ohms before saturation would be an issue. Nowhere near the 1/2W rating.
One thing that's been on my mind regarding the Edcor's, is the max input voltage. They list 15vrms for the PCW series. From what I've been simulating in Spice, it seems I can only get 50-100mW into 32ohms before saturation would be an issue. Nowhere near the 1/2W rating.
Unfortunately, there seem to be no other options open for headphone amplifiers - Transcendar used to carry a single-ended transformer meant for headphone duty (reasonably priced, too), but they are only re-winding transformers now, not making new ones. I got while the getting was good, and secured 2 pairs of the gapped transformers, and one pair specially non-gapped for parafeed use.
Maybe it's time to haul out the pair of headphone output transformers that Jack at Electra-Print wound for me (with lots of curmudgeonly kvetching) and actually characterize them on the bench. They are rather large for their intended application, so maybe he included some reasonable primary inductance as part of the bargain. I had intended those transformers a part of a feedback-based amplifier utilizing a 5965 dual triode, but the project failed due to stability issues. Maybe a 6GV8 SE design with global feedback will work out better.
Last edited:
I just got 8 pieces of the Russian equivalent of the 6GV8 in the mail today, so I might start designing a circuit. Part of it will depend on how the Elactra-Print transformers measure up.
OK, I plunked the oh, so reluctantly wound Electra-Print headphone transformers on the bench for evaluation. With 30V, 300Hz excitation, they measure exactly at 3k:35 ohms turns ratio. They also have 15H of primary inductance, so they will likely need some negative feedback to prop them up at the low end. Primary current capability is not certain, but with only 15H inductance, the gapping will probably support a fairly substantial primary current.