February 2015

Bottlehead The Quickie Vacuum Tube Preamplifier
A $99 battery powered directly heated triode preamplifier kit.
Review By Jeff Poth

 A long time ago (in a galaxy... -- Ed), some of my first building experience (and much learning) was with the Bottlehead Foreplay preamp, which was $99 excluding base for the stock model, and had some upgrades available. I spent many hours modifying it, tube rolling, trying different attenuator options, different power supply arrangements, chassis damping, and pretty much every which way but loose to mess around with the thing. Part of my challenge was that I was using a high input sensitivity power amp with relatively high efficiency speakers -- a situation which persists today, albeit at a much higher level of performance than way back when. A few years ago I reviewed the Bottlehead SEX amp, and was thoroughly impressed with its performance, a musical, flexible, and enjoyable small integrated amp.

In late 2014 it was time to bottle things up again and get some more goodies, and Bottlehead sent along review kits for the Quickie and the Quicksand (a 4WPC battery-powered class D amplifier). These are their two least expensive "active" components at $99 and $119 respectively ($20 off ordered together) but that's not a problem for me; I love a good bargain and simple circuits can have a charm all their own. One of the charms of these particular bargain components is in the top plates supplied; they can range in color based upon what the acrylic supplier sends (a cost-control measure). I got red for both plates, and I rather like the look of them (seen here).

The price of

Quick Quickie Overview
The Quickie uses one directly heated 3S4 pentode per channel, wired as a triode. The difference between indirectly heated and directly heated tubes is that directly heated tubes pass current through the cathode of the tube to heat it, where indirectly heated tubes have filaments used to heat an electrically isolated cathode. Directly heated require individual heater supplies for each tube; these must carefully be designed as the heater supply will also define some of the amplification characteristics. Directly heated tubes tend to be more microphonic than their indirectly heated cousins as well, and indirectly heated tubes are a little easier to design with as the cathode can be managed independently of the heater supply making noise management a fair bit easier. Why then would anyone want to use directly heated triodes? Simply put, they're believed to sound better. The vast majority of Single-Ended Triode (SET) amplifiers are based around a handful of tubes (300B, 2A3, 45, 845), all of which are directly heated. The battery power means that having the individual, low noise supplies for the heaters is no problem, which is part of why the Quickie "works".

The Quickie circuit is the simplest you're likely to see -- a grounded cathode amplifier or "plate amp". This is one of the key building blocks of most tube circuits, and is present in most tube amplifiers in some variant or another. John Broskie of Glassware fame discusses this topology here.

The 3S4 is a pentode, wired as a triode in the Quickie and has a center-tapped cathode which runs with the two sections in parallel off a 1.4V 100mA supply, they can also be run series with 2.8V and 50mA. D cells give 1.5V when fresh which is a little hot but forgivable, and as they drain the voltage falls in a fairly linear fashion. You'll want to monitor voltage from the D cells (one per channel- dual mono heater supplies are necessitated by DHT operation); I have taken to checking it upon power up each time I listen. The high voltage.. (choke... higher) voltage supply comes from a quad of 9V batteries wired in series for 36V with alkaline batteries.  Bottlehead indicates about 100 hours of battery life, I found that by about 25 to 30 hours of playback I had dropped the Duracell D cells to around 1.25V, which I deemed to be low enough to swap in fresh, but this may be on the low end of cell reliability, some other brands have seemed to fare better but I haven't tracked their timing as closely. Shutting down overnight will often recover a little voltage, too. During that same stretch the 9V Duracell Procell alkaline batteries I initially used dropped to about 8.35V per; not surprisingly they retain a longer life than the D cells which draw several times more power (100mA per cell @ 1.5V compared to about 4mA at 9V). Longevity will vary with brand and type, so your drain rate may be slower, or faster. Do buy your batteries online and Google the specific types you'll be getting, as you want high mAh (milliamp hours) batteries from a reputable company (that's not only Duracell or Energizer btw). Batteries are a high markup item at most retail outlets.

So, batteries come with two major drawbacks: additional operating expense and ongoing maintenance to ensure that you're getting appropriate voltage. Apart from portability batteries have two key positive aspects to their performance: they're typically quiet and relatively low impedance, both of which are very good things for a power supply. There is some variance in noise performance and impedance within the different cell types; from what I've seen good quality Nickel Metal Hydride (NiMH) rechargeable batteries having some of the best noise performance, with Alkaline batteries acquitting themselves very nicely but Nickel Cadmium and Sealed Lead Acid both potentially being a little noisier. I haven't found definitive noise testing data for batteries however.

More On Batteries
I wanted rechargeable batteries to keep the cost managed, and what I found was a little problematic in terms of getting rechargeable batteries that would play nice with the Quickie. First, because NiCads have a memory effect and appear to be noisier than Alkaline or NiMH, they are a non-starter (for me). This leaves (as reasonable options, there are other technologies/chemistries) Lithiums, which have a pretty specific care and feeding regimen, and Nickel Metal Hydride. One key advantage of better quality rechargeable cells is a flatter discharge curve- they retain a relatively stable operating voltage up until they approach the end of their energy storage. This allows for more stable operating points within the preamp. NiMH have a couple drawbacks though, first is that the chemical cell is typically good for just a little over 1.2V. Typically alkaline 9V battery are actually six small Alkaline cells in series, as typical Alkaline battery chemistry gives 1.5V. When a NiMH 9V battery is constructed in this fashion, only 7.2V are available (this is fairly common practice). There are 7 cell and 8 cell 9V types, giving 8.4V and 9.6V respectively; much better if you're worried about too low a voltage. The other issue is that many NiMH cells have a high self-discharge rate, thus they're not shelf stable. That means even if you don't use the preamplifier (say it were in a garage system, or some such), you'd still come back to substantially drained or dead batteries after a few months. Not such a biggie if you're willing to maintain the charging cycle, but there are some modern low self discharge (LSD) NiMH that reduce this issue significantly.

For the 9V batteries I wound up with 9.6V Imedion from Maha. They are pricey but high quality and the Quickie can handle extra voltage (and indeed eats it up). In circuit, fresh off the charger the Maha give 11.3V, driving the B+ up to 45V. The Quickie can handle the excess voltage and seemed to have a little extra pep in its step with this higher voltage. With the constant current supply in place, the plate voltage was about 31V and cathode was biased to 2.5V. After 5 to 6 hours they'd settled down to 10.3V, still a significant, and welcome, bump on the "high" voltage rail. They stay a little over 10 for quite a few hours, I haven't done the formal testing with them as I picked up 8 and just swap them out every once in a while so I never have to test nor worry about the 9V cells. The higher voltage than nominal is probably due to the low current demand- good high quality NiMH handle heavy loads much better than Alkaline (less voltage drop), and so at a handful of mA, they're loafing along and exceeding nominal voltage. The higher plate voltage also runs a little more current through the tube- not a bad thing (within reasonable limits).

D cells are a bigger challenge -- 1.2V from NiMH won't cut it, and 3V Lithiums are too much (one could regulate them down, or run the heaters in series instead of parallel if desired but that's another ball of wax). There are Nickel Zinc and RAM options, but neither are great. I am sticking with Alkalines and a replacement cycle, for the short term.

Basics And Assembly
In the Quickie, there are only two controls- the selector switch which doubles as a power switch and the potentiometer. The tube circuit is only 3 resistors and 2 capacitors, but if one chooses to add the current source upgrade, the complexity shoots up a bit more- still pretty minimalist overall though. Assembly is incredibly simple and the pdf instructions provided made it a complete cakewalk, particularly with a tablet display handy to refer to. Everything down to lengths of wire are specified and it's something that anyone who can solder without getting blobs everywhere can put together with ease. After assembly a simple resistance check then a voltage check with a multimeter at several key points within the circuit confirm that the preamp is ready to rock. A total of about six hours work went into my build- about four for the circuit and an hour for each the base and the constant current source build/retrofit. The base doesn't even require clamping since blue painters tape works great, with care.

Some light grousing- I had glued up both of the alder bases from Bottlehead, which are the same size for the Quickie and the Quicksand, and let the Quickie work from the unfinished base while I put a nice Danish oil finish on the other base after a 220/320 sanding- higher grits and more sanding are advisable for a better finish, but I didn't want to go too far off the plans on this step, which call for only sanding to 220. After letting the base cure, I went to swap in the finished base for the unfinished, and had a rather rude awakening- the base was slightly smaller than its brother. I had to then trim a little extra length into the rabbet on an assembled and finished wood base, which I did with a knife and some care, but then had to re-finish that portion and whatnot- one of the few negatives I have to gripe about, the other being some chipping around the cuts on the acrylic plate. Neither were major issues but when you buy an inexpensive kit, some of these challenges are almost inevitable. I do advise test-fitting your top plates carefully before applying a finish.

Performance
Once the preamp was done (before the constant current source was installed), I hooked it into the big rig- a risky proposition for any preamp, as my rig utilizes high power/high gain amplification and big, high efficiency speakers, see here. The first thing I noticed was that there was a low volume high pitched singing coming from the speakers (maybe 55 to 60 dB). This died down after a moment, but upon adjusting the controls, either upon the preamp or any of the connected gear, I got the same ringing at about 3 kHz. This was tube microphonics as the 3S4 is known for that drawback, and indeed it's well-documented as a challenge with this pre. Because it was at a level that could be overlooked temporarily, I proceeded to run some tunes through the preamp, with a 50k amp load (2x100k amp channels in parallel).

Now, we're talking- the Quickie was going into a rig which required high attenuation from the preamp, making it a major challenge for any but the quietest preamps. The Quickie, apart from the microphonics, is very quiet and tube self-noise is the only other issue you may face with some tubes, and shouldn't be an issue in all but the highest gain/big amp systems. The ringing noise, with all but one tube I tried was audible and excited acoustically during music playback, but it is fairly benign.

One of my favorite bits of music to evaluate a new component with is Trevor Pinnock with "The English Concert" doing Handel's Water Music. This is a lush, rich recording that is tonally robust without being overwhelmingly ambient. The Quickie didn't disappoint as I settled in and was able to relax, close my eyes, and focus primarily on the music. Any errors the Quickie was making relative to my transformer attenuators (which are quite pricey, and worth it) were primarily errors of omission. I was giving up some air, resolution, definition, and a meaningful chunk of soundstage width; but what I wasn't giving up was any meaningful amount of musicality or dynamics with this recording. Some more challenging recordings did indeed tax the dynamics a little bit when going against something as nimble and dynamic as my TVCs, which are exceptional in that regard. In short, the Quickie, even before the addition of the CCS, performed quite well, with some shortcomings but a relatively clean presentation, with plenty of musicality and excitement.

Going through a variety of other recordings, I found that at higher volume levels there was some added glare relative to my reference transformer volume controls. Some of this seemed to be microphonics rearing its head, and the solution for this is often to swap tubes out to find a low microphonics pair. The PJCCS (more later) alleviated most of the perceived glare. The Tung-Sol that shipped with the pre were nice tubes, but I ordered some alternatives, NOS RCA and Mullards to try some alternatives and see if I could minimize microphonics. The microphonics issue seems to be more related to the excitation of an internal structure than to surface resonance, so simply damping the tube envelope will only help somewhat. The sockets can be mounted with O-rings to provide some damping, and there are other methods to mitigate resonance getting into the tubes but this review is focused upon a stock preamp. If you have a high sensitivity amp/speaker combo, some ringing is something you'll have to live with the stock Quickie, plain and simple. Out of 10 tubes I tried, from Tung Sol, RCA, and Mullard, only one stood out as being less microphonic than the others. Self noise was only a very minor issue, in this challenging system, with one of the ten having a whine at about 15 kHz that was loud enough (in this system) to be mildly problematic.

Adding ThePJCCS
The PJCCS is the "Paul Joppa Compensating Current Source", a current sourcing transistor circuit to be added to the power supply to linearize the tube behavior, particularly as the battery voltages in the stock kit (when using Alkalines) fall. This counteracts some of the distortion-inducing current variations, and increases power supply isolation- this last part is important. The PJCCS was an easy affair to assemble though it's a small circuit board and so needs careful attention and good eyes or magnification. It replaces the plate load resistors in the circuit, so those are removed before installing this board. My results were quite positive; the CCS gave me a more dynamic sound, with better resolution, image definition and a broader soundstage (which I believe to be due to the additional power supply isolation/separation). Treble was to be better represented, with improved tonality of small percussion instruments, and a greater sense of naturalness and ease of presentation (which included elimination of much of the higher-level glare I experienced with the stock layout).

In short, the CCS is not optional if you want to get the cleanest sound out of the pre. Some will prefer without, enjoying the stock colorations, perhaps finding it more pleasant. I prefer the CCS in place and would go so far as to say it's significantly better and more accurate that way. While the addition of the CCS didn't get it to the level of the reference TVCs, it did bring the pre much closer, and that's no small feat. The transformers, raw, cost many times what the Quickie does even before switches, connectors or casework.

Some Other Systems
After enjoying the PJCCS, I decided to give the Quickie a shake in a system with a little less input sensitivity, and indeed, when the Quickie is in a system better matched to its gain structure, (meaning you run it at higher volume than the extreme low end of the volume control my big rig demanded), the microphonics and self-noise from one singing tube were somewhat less problematic, while the rest of the preamps sound remained very solid. This system was my daughter's bedroom rig, using the Zigmahornet cabs loaded with modified full-range drivers (utilizing planet 10 hi-fi phase plugs).

I then decided to add in the subs in the main system, driven in parallel with the amp inputs forcing the Quickie to drive three channels of amp per output. This means a 9700 Ohm load or so for the Quickie, it was my hope that the sub amp input would act as a current sink for the ringing, and indeed, it did; the microphonics issue subsided a bit and became less of an issue in my rig. Unfortunately the lower load along with some extra cable capacitance (presumably) collapsed the soundstage back down. Within the more limited soundstage tonality remained good and dynamics were okay, but the Quickie was definitely happier into the higher 50kOhm load as it gave up a fair bit of resolution, space, and basically every audiophile term into this load. Cable capacitance wasn't out of the range of normal, being equivalent to about a 2.5 meter run of typical twin-axial audiophile cable. The good news is that it remained musical and enjoyable- the things that matter most.

So, making sure that I give this pre a very fair chance to compete in a system more appropriate to its cost, I combined it with the Bottlehead Quicksand amp, the other battery-powered Bottlehead device I was sent for review. This combination, in my daughter's rig vs. a modest Audiosource A100 integrated amplifier was a huge upgrade, with two caveats. First, the microphonic ringing was still an issue and is audible between notes. It's pretty bandwidth limited to the resonant frequency so it's not a horrible thing, nor is it extremely loud, but it's a problem. The second was that her CDP didn't like an ungrounded pre/amp combo and so there were mains hum and buzz. Using a portable source cured this, since the inexpensive Sony player was the culprit. I didn't experience any of this noise when using the Quickie in the big rig. Some systems may be problematic, though. But all that aside, the sonic quality of the Quickie via the Quicksand was head and shoulders over the more powerful Audiosource, exhibiting a more natural tone, better imaging, and a more musically enjoyable presentation overall.

Conclusion
Stock, both with and without the CCS upgrade but particularly with, this is a nice little linestage. Doc Bottlehead's $99 Quickie vacuum tube preamplifier is enjoyable to listen to, fun to build, plus has a lot of visual and technical personality. Yet of more importance is that the Quicky preamplifier produces a surprisingly musical presentation without going all ooey-gooey as some bad tube circuits can. I can wholeheartedly recommend this kit as it's a hoot to build and listen to. It is something that you'll want to play with if you're a DIYer / tinkerer. It won't replace a very high performance linestage in hi-fi terms, but the sound is enjoyable and compelling- it errs in the ways that seem to matter less to the music and more to... ahem... reviewers. Assembly is easy if you can be careful and solder cleanly, and the design is well thought out. There are some challenges as the batteries can be a pain and/or pricey, and the microphonics issue will take some working out. I am looking forward to an ongoing adventure with this preamplifier. There's a huge amount of potential here, and even with the simplest stock version, it is fun and enjoyable.

Specifications
Type: DIY stereo vacuum tube preamplifier
Price: $99 plus a variety of options to choose from.

Company Information
Bottleheadquarters
9415 Coppertop Loop NE
Suite 101
Bainbridge Island, WA 98110

Voice: (206) 451-4275
Website: www.Bottlehead.com