Coming soon to a workbench near you…

In a world where near-field and headphone listening has become an unstoppable force. Where every DIY builder is bored to death of rational and safe two-stage, single-ended triode designs. Where power supplies have become an afterthought and parts values are just plugged and chugged. Prepare your butt for a new madness. Prepare it for La Luciérnaga…

the-big-sheet

Note db scale for the predicted frequency response graph: +/- 1 dB from 40-20khz. If that’s not good enough, I’ll show you how to make it even better. There’s mucho tube math coming your way, amigo.

Letters to WTF: All-tube MC phono preamp?

Q:
 ­
Why not build an all-tube MC phono preamp?
A:
 ­
I haven’t built an all-tube MC phono preamp.  I build MM stages with the assumption that MC users will use step up transformers. Generally, its much easier to keep the whole thing quiet that way. This is a simplified explanation, but I hope it gets the general idea (SNR) across:
 ­
Tubes impart noise in a few ways but they’re all usually tiny.  MC carts need around 60-70db of gain to bring them up close to line level.  This is multiplying the MC signal from the cart by 1000-3000x before it gets to your amp. In contrast, a MM cart usually needs 40-50db, which is an amplification of 100-300x.**
 ­
The more tubes used, the more very tiny sources of noise get introduced.  Although the noises are tiny, they are amplified by the preamp, just like the signal. So the more gain required to get the signal to a desired listening level, the closer the tiny noises get to being audible as well.  Several stages of tubes for a very large amount of gain can lead to unwanted noise for this reason (guitar amps and phono preamps are both good examples of this).
 ­
The step up transformer cannot practically do more than about 20db-25db of gain without having some undesired consequences, but they don’t have ‘moving parts’ and are very good at rejecting noise/hum.  The 25db of gain from a transformer is enough to lower the tube gain needed from 3000x (70db) to 175x (45db), meaning any noise from the tubes is amplified much less. It’s kind of like giving the tube part of the preamp a head start in the race against noise.
 ­
So that’s why I haven’t built a MC phono preamplifier. If I were to build one, I’d most likely add step up transformers to a MM preamp like the El Matematico or similar.  To me, this is the most practical approach with the highest likelihood of success.
 ­
 ­
**At the risk of muddying the waters, phono preamps actually need about 20db more than the numbers mentioned here but this is then attenuated by the RIAA correction filter to result in ~40db or ~60db net gain, MM and MC respectively.

New Page: Shunt Regulators

shunt-reg

I’ve been doing some reading on tubes in shunt regulator power supplies lately (lots of great articles on TubeCAD including this one).  I’m planning to incorporate one in an upcoming build.  In operation, this isn’t too different from the VR regulator power supply in my Matemático Phono Preamp, but a shunt regulator with a triode would have an adjustable output and might afford even better ripple rejection.

My recent series regulator project is another example of power supply regulation.

Click here to see the new page on shunt regulators.

Something for beginners

Pete Millett’s Starving Student was one of the first amps I ever built completely from scratch. Unfortunately, the 19J6 tubes have become rare (or at least no longer dirt cheap) due to all the bright eyed DIYers scooping them up to build amps. I think the world needs another <50V tube amp for beginners, so I’m designing one. Like the original, it’s an oddball tube with a MOSFET buffer and an off-the-shelf power brick (same brick, in fact).

Millett is one of my personal tube heroes. This is a tribute.  Full write up coming soon (and parts values subject to change once tested).

 

Letters to WTF: rectification

Q:

What is your opinion on rectification?

A:

Tube vs SS rectification is a hotly contested topic in the audio world and what I tell you will be opinion.  That said, I’ve used both and never had a problem (eg here, here, here, here).  When designing from scratch, my choice is dictated by my power transformer, space available on the chassis, and whether a tube rectifier is in the budget. Sound quality is not really a consideration.
­
 ­
Of all the parts in a Class A single-ended amp that contribute to the sound, I think rectification is one of the least important. If you end up with the right B+ voltage and you filter it so that there is no ripple, you’re 99.9% of the way there. The debate about tube vs SS rectification usually focuses on theoretical advantages or disadvantages without respect to whether a supply is well-designed and implemented.
 ­
All things being equal, and space and budget allowing, I tend towards tubes because I like the way they look and I like knowing that the high voltage will start up just a little bit slower. All things being equal, others would go for solid state because it has a lower output impedance and allows for larger filter capacitors.
 ­
There’s more emotion than measurement in either decision.
 ­
274B borrowed from thetubestore.com and 1N5408 borrowed from PartsExpress.com. I claim Fair Use here and anyways they can get bent if they don’t like me linking to them in exchange for the pics.

An open letter to transformer manufacturers…

Dear transformer manufacturers,
Here’s a quick list of existing products that I have found with comments on their suitability for the DIY market. This is followed by blue-sky ideal output transformer specs.
 ­
Edcor PCW10k-7k/300-32: this is a nice low cost but suffers the same issue as all of Edcor, namely lead time and lack of availability overseas.  The turns ratios and secondary taps are just about ideal for headphone applications (which I believe it was originally intended for).  The board mount style, on the other hand, is less than ideal for most hobbyists.  With a rating of 0.5W, 15Vrms on the primary, and no published inductance figures, it also hasn’t really seen much interest from a lot of the community. That said, it’s the only non-custom North American option I know of. This is non-gapped single-ended so it’s limited to parafeed applications. Honorable mention to Edcor’s WSM and XSM series, though.
 ­
Sowter options: by all accounts these are very nice non-custom options, but they are difficult to get in North America and the price is very high. The way the secondaries are wired also makes switching or wiring multiple jacks difficult. Because most headphone enthusiasts have multiple pairs at different impedance, using an output transformer with an easily switchable secondary tap is preferred to ensure consistent load reflected to the output tube. The Sowters don’t make this easy, unfortunately. Most of the Sowters are gapped for normal series-feed single-ended power. End-bells are nice.
 ­
Lundahl LL2765: this is a fancy c-core option with a secondary arrangement similar to the Sowters. The same caveats about easily switching the impedance configuration apply. The turns ratios and impedance options are close to ideal for the headphone market.  I haven’t been able to find a price (Lundahl is distributed by K&K Audio in the USA, though LL2765 isn’t on their price list). I expect it’s in the $200+ each range. No end bells, so no fun in showing off the iron. It’s available with a gap (series-feed) and without a gap (for parafeed or push pull operation).
 ­
As a DIY headphone amplifier designer and enthusiast, what I’d love to see is an output transformer that is readily available in North America and Europe at an affordable price.
 ­
Headphones broadly fall into impedance categories of 200-300 and 32-75 ohms (with most recent products trending towards the lower impedance). I think secondary taps of 200 and 50 ohms would make sense as these would be a ratio of 4:1 for full winding and center tap, easily selectable with a switch. The center tap secondary, rather than asymmetrical taps, might make this transformer useful for other balanced applications as well (and if so, might as well put a center tap on the primary, too).
 ­
A primary impedance of around 8k would work well for a good range of tubes. If rated for 8k:200, we’d have a turns ratio of about 6.3:1. With 300 ohm headphones it would provide a 12k impedance and with 150 ohm headphones it would provide a 6k impedance. If using the 8k:50 center tap, it would provide a turns ratio of 12.6:1.  That’s about 5k with 32 ohm headphones or 12k with 75 ohm headphones. All these numbers would work fairly well with small popular single-ended tubes (EL84, 6V6, 6S45Pi, etc).
 ­
The transformer would ideally be rated for 2-3W.  Headphone users are as silly about their power as speaker users are.  Never mind that they only need a few mW to reach 100db. Frequency should be rated 20-20k +/- a db or two or the transformers would be ignored by spec snobs. More similarities to the rest of the high end audio market there.
 ­
Gapped and rated up to 50mA would be excellent, but not if it compromises the frequency bandwidth through low inductance or the cost by making it huge. A non-gapped transformer for single-ended headphone amps works a little better than for speaker amps because we can use it as parafeed, keep the B+ voltage reasonable, and still make plenty of power. Non-gapped and an exotic metal core would be even better, but I don’t see it much in most transformer catalogs.
 ­
So, TL;DR, what I think would be ideal is:
– 8k:200ct impedance
– 2W+ rated output
– 20-20k rated bandwidth
– $100 per pair (less is better)
– option for end-bells
– gapped (~50mA if possible) or un-gapped
 ­
Turning that into a small product line, I could see it working out as something like:
– 2W gapped, no-end-bell
– 2W gapped, end-bell
– 4W un-gapped, no end-bell
– 4W un-gapped, end-bell
 ­
I’d imagine the cores for the above would be around similar sizes, so materials and winding are more or less consistent, but the more in depth transformer permeability math isn’t something I regularly do. The only differences would then be in adding a gap and/or adding end-bells, making a pretty efficient line in terms of manufacturing and materials.
 ­
As you probably know, the headphone market is growing pretty quickly.  Perhaps not growing as fast as the turntable market, but the overall size is pretty huge as of now and is still on the uptick. I think the high-end headphone products category will continue to grow. Most headphone companies are trending upwards in price points and releasing more designs that are meant for home-use (and maybe that’s partly driven by the growing turntable market, too). I don’t think I’m the only one waiting for a good practical headphone output transformer from an established and well-distributed manufacturer.
 ­
So who will step up to the plate?
 ­
Sincerely,
WTF Amps