|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
97.105.99.138
I've been thinking about re-purposing some old consumer-grade SE output transformers that are just collecting dust. One of their potentially useful purposes would be to connect just the primary winding in a choke-loaded cathode follower stage, leaving the secondary unused. Is it reasonable to expect a 5K:8 transformer designed for use with a 6V6 or 6BQ5 in a consumer hi-fi application (Magnavox, Stromberg, Zenith, etc.) would exhibit at least 60-70H primary inductance?
--------------------------
Buy Chinese. Bury freedom.
Follow Ups:
Paul and hennfarm, let's say the primary is only 10H. I don't know the typical DCR, but I'll check this evening after work. For now, I'll assume 500 Ohms. If I drive a signal across that inductor using a generator with 100 Ohms output Z (similar to the followers I have in mind), 20 Hz response is down only 0.26 dB relative to 1 kHz. If the inductor is 20H and 1k Ohm, 20 Hz is down only 0.13 dB. This isn't as good as 50H or larger, but I'm still thinking the concept might be viable.
--------------------------
Buy Chinese. Bury freedom.
At 20Hz, 10H is a reactance of 1250 ohms. A signal of 100vRMS would cause an AC current of 80mARMS, or peaks of +/-112mA. The choke must be able to take that current in addition to the quiescent DC without saturating, if you are to avoid added distortion.
"The choke must be able to take that current in addition to the quiescent DC "
Paul, I think these consumer-grade transformers are capable of much more than that. They're designed to work with a SE pentode like the 6BQ5 or 6V6 drawing 60-70mA at idle. In this service, the primary winding would be subjected to at least 160V RMS to produce 5W at the secondary. My application is much less demanding in terms of voltage and current, probably only whatever is present at the cathode of a 7044 or 6BX7 direct coupled to an output grid.
What really interests me about all this is the increase in performance of the winding when used in a low power, low-Z circuit. I believe both the high and low ends will be extended, relative to the original application. Incidentally, I picked out a transformer this evening, and the primary DCR measures 450 Ohms. I'll test it for inductance this weekend.
Honestly, I'm excited about the prospect of using these to choke load CFs. I've been looking for a less expensive alternative for some time. :)
--------------------------
Buy Chinese. Bury freedom.
OK, I was able to measure the primary of these 5K:8 transformers at three points -
20 mADC: 20H
40 mADC: 14.5H
60 mADC: 10H
I'll have to wait for an actual CF project to test frequency response, but I expect it to be smooth and extended in such a low impedance circuit. Great, it looks as though these will work well with nearly any CF driver!
--------------------------
Buy Chinese. Bury freedom.
Keep in mind that low impedance output impedance is a small-signal value. A cathode follower can deliver the most power into about the same impedance as a plate-loaded stage. Available undistorted output swing will be less as load impedance drops.
Yes, I understand the load shouldn't be a lower value than if it were located at the anode. I'm making use of the lower Z (and reduced DCV) at the cathode solely to permit direct coupling. Eliminating the coupling cap is primarily for the purpose of ensuring near immediate recovery from output stage overload.
--------------------------
Buy Chinese. Bury freedom.
Not if it's a single-ended transformer. For example, the Hammond 125ESE is popular, and bigger than most consumer-grade transformers, which should enable it to have more inductance than the usual Magnevox et al units. The Hammond catalog page shows a measured 10 henries, for a nominal 5K impedance level.
Some 15 years ago I measured the ungapped 125E, which is smaller. It managed 22 henries with zero DC, falling to 18H at 12.5mADC, and 11 henries at 25mADC.
I have also measured the larger of two transformers sold as replacements for the Fender "Champ" amp (SE 6V6). Rated 5K primary, inductance was 8.7H at 45mADC. It was a remarkably sweet sounding transformer, as long as you don't feed it bass.
Sorry TK, more like 10henries, for the manufacturers you quoted. I've measured and sifted through a lot of chaff in my diy journey :+). JH
Edits: 10/13/16
When you guys measured the primaries, did you apply a DC current? Is it possible the numbers in practice might be worse than what you saw?
--------------------------
Buy Chinese. Bury freedom.
My measurements are made with a variable DC current plus about 12vRMS at 60Hz.
FWIW, I figure that when the current is enough to reduce the inductance by 20% relative to zero DC current, that's a practical maximum-current specification for output transformers and plate chokes. I base this on my measurements of a fairly large number of commercially-available SET transformers and chokes.
Measured inductance will increase if a higher AC voltage is applied, or if a lower frequency is used, as long as the flux is below the saturation level. A higher frequency (such as 1000Hz, widely used in multimeters) or a smaller voltage will give a smaller inductance.
another important thing to note is that meters that sample at 1Khz often give incorrect results since the self resonant frequency of many of these devices is in the 1Khz range.
dave
Many thanks, Paul. You did talk about that in your earlier post. I got distracted at work and forgot I'd read that part before replying. Old age, I guess. Anyway, this might be a good time to use my CCS test method for checking the transformers with various DC currents. Current in the final circuit should always be much less than the SE stages these transformers were designed for. Thanks for the info, I'll report back with results in case anyone else is interested in trying this.
--------------------------
Buy Chinese. Bury freedom.
I use an Amprobe Lcr55a (1khz) with no DC current through primary. From what I have read and what Mr. Joppa has already stated inductance falls with DC current through the winding. Every device will be different, some may have an impedance hump <100hz.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: