• 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.

Beginner DIY Amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
In a tube amp build you will find that the iron and chassis are the most expensive parts (unless you are running some truly exotic output tubes). Even going with (the upper-level of) Edcor iron and a cheap Allied power transformer, I ended up spending ~$750 to build this single-ended pentode (ok, beam tetrode) amplifier.
6th Street Bridge: Project: The Command 1625 tube amplifier

Top plate was done by Front Panel Express. Output tubes are the cheap and plentiful 1625. It might make 10Ws of power with a stiff breeze blowing.

Back when Dynaco 70s were cheap and plentiful this would be my first recommendation for messing with tube amps. These days? It's hard to say. I've had good luck buying singles of Heathkit monoblocks - like the W-2, W-3, W-4M, and the UA-1 or UA-2.
 
Last edited:
I think it's really hard to beat a push-pull EL84 amp for most bang for the buck.

There are several good designs here on diyAudio.

Baby Huey
- Baby Huey PP EL84 amplifier - diyAudio

Eli Duttman's El Cheapo could be adapted to EL84 with distributed load connection, which should make over 10W per channel, cleanly.

Red Light District (more advanced)
- http://www.diyaudio.com/forums/tubes-valves/81492-red-light-district-another-pp-el84-amp.html
- The Red Light District: A 15W Push-Pull Amplifier

Also, there's a PP EL84 amp in Morgan Jones' "Valve Amplifiers" 3rd Edition, called the "Bevois Valley." That's a quite competent amp, and fully discussed in his book.
- https://books.google.com/books?id=e...G#v=onepage&q=bevois valley amplifier&f=false

- http://www.diyaudio.com/forums/tubes-valves/64110-morgan-joness-bevois-valley-amp.html

__________________________________________________

Why push-pull EL84?

1) The output tubes sound good, and they're relatively cheap, about $12 to $15 each (four required for stereo).

2) EL84 is easier to drive than just about any other output tube type. 12V peak signal out from the driver is all that's needed. (24V peak drive needed for -6dB negative feedback.) A 12AT7 or 6DJ8 can do that with few problems.

3) The EL84 runs on a relatively low voltage of about 320V DC (maximum), which means you don't need 450V or 500V rated capacitors (which are expensive). 400V rated parts will do fine.

4) The tubes are small. The chassis doesn't have to be large to provide sufficient cooling and room for wiring.

5) There are several decent output transformers available for this kind of project. The Dyna Z565 remake is supposed to be good. Edcor makes a 10W rated OPT that actually works pretty well, and is cheap, the GXPP10-4-8K for $28 each -- https://www.edcorusa.com/gxpp10-4-8k. That one's with a single 4 ohm secondary. You can also get it with 6 ohm or 8 ohm secondary.

Just a quick look at the cost of the various major parts...

- 4x EL84 = $60
- 2x 6DJ8 = $25
- chassis = $50 or so
- 2x Edcor 10W OPTs = $70 (incl shipping)
- 1x Edcor 500VCT 200mA power xfmr = $70
- capacitors, say $30 (?)
- resistors, say $20 (?)
- 6x 9-pin mini tube sockets = $30
- 2x RCA jacks (insulated) = $5
- 2x dual 3/4" spacing binding posts = $5
- electrical cord, strain relief bushing/grommet or IEC inlet, etc... $10 (?)

So far that's $375 already. This is not a cheap hobby. But I was conservative with the pricing of parts above. You can save money by shopping for parts at places like Apex Jr.

If you don't already have the tools for making holes in your chassis, and the other hardware, then you'll be adding that to the cost of the project.

All the best.

--
 
Last edited:
Anytime an amp incorporates a NFB loop that includes the O/P transformer, it is essential to have magnetic headroom. The headroom accommodates the low frequency error correction signal. Without headroom, the core of the "iron" will saturate amd the amp will sound wretched. Amps built with "12" W. tubes, like the 6V6 and EL84, need O/P "iron" rated for 25 to 30 W. TANSTAAFL will forever plague us in this hobby.

An "El Cheapo" can be built using EL84s easily enough. The O/P tube cathode bias resistor changes. Ask Jim McShane for the correct value.

The 6CM6 is a 9 pin miniature (Noval) based member of the 6V6 clan, as is the 12AB5. 6V6 clan members draw less heater current than EL84s and they happen to make better triodes. ;) Both the 6CM6 and 12AB5 are out of production and several sets would have to be acquired. OTOH, those types are very inexpensive. :cool:

Great flexibility can be obtained by building with Octal sockets and constructing adapters for the other basing types. Switches to wire "6" V. heaters in series and "12" V. heaters in parallel is another flexibility possibility. The 6V6 clan is large and includes: 6AQ5, 12AQ5, 7C5, 14C5, and 12V6.
 
Last edited:
Eli D wrote:
Anytime an amp incorporates a NFB loop that includes the O/P transformer, it is essential to have magnetic headroom. The headroom accommodates the low frequency error correction signal. Without headroom, the core of the "iron" will saturate amd the amp will sound wretched. Amps built with "12" W. tubes, like the 6V6 and EL84, need O/P "iron" rated for 25 to 30 W. TANSTAAFL will forever plague us in this hobby.

Yes, I stand corrected as far as PP amps with global NFB are concerned. :)

One way to get around that little problem is to bandwidth limit the input to within the capabilities of the output transformers. For instance, with a 100k pot at the input, a 0.1uF cap would impose a -3dB point of about 15Hz. It would probably be down -1dB at about 30Hz or so. It works, but the trade-off is that you don't get deep bass from your amp. Then again, why would you expect a weenie 10 watt-per-channel tube amp to play deep bass, especially if your speakers are 2-way bookshelf types that don't go any lower than about 50Hz? Frank Van Alstine used to bandwidth limit both the low and high ends of Dyna ST70 amps, and even though it's not a popular thing to do, it does clean up the sound of a typical UL amp with global NFB. Of course there's no free lunch if you put this high pass filter at the input... You lose the low frequencies below 30Hz or so. But do you actually need those?

Oh yes, and... The Baby Huey amp runs plate-to-grid local feedback around the EL84's (not around the OPT). Ian (its creator) said he runs his without global NFB. I guess that means you can use a less capable OPT, maybe. Cheapness! However, the BH does require a negative voltage supply for the 12AX7 input tubes. That adds expense. But before you give up on the idea, I improvised on Ian's theme and came up with a version of Baby Huey that doesn't include any transistors. That could be a cheap way to go... Bear in mind that Ian meant Baby Huey to be a 'poor man's SE 300B.' He was going for a particular harmonic signature, so it's not to be thought of as a low distortion design. It's just meant to sound, er... 'musical.'

If you must have deep bass from a little tube amp (and with lower distortion), you get to spend more money. In Edcor iron, the choice would most likely be the CXPP25-MS-8K/23% at about $59 each, which gets you a multi-tapped secondary (4, 8 and 16 ohms) and 23% screen taps, which some say is the correct tapping for ultralinear operation of EL84 pentodes. (You'll probably end up using the 4 ohm secondary tap with most contemporary speakers.)

The Dyna Z565 clone from Triode Electronics costs $89 each.

The Transcendar version of the Dyna Z565 is $75 each. That one has 25% screen taps.

6V6 tubes are not often run ultralinear. I haven't read anything definitive about whether 6V6's should be run with 40% or 25% screen taps for ultralinear operation. Anybody know about this?

Maybe the cheapest way out is to run some sort of 6V6 variant in triode, and don't use global negative feedback around the output transformer. Choose as high a primary impedance as you can get, to increase damping and lower harmonic distortion. Then perhaps you can use OPT's that are meant for lower output, which means they'll be physically smaller, which usually means better high frequency response. This cheapie transformer looks interesting -- 10 watts, 12k ohm primary, 4/8/16 ohm secondary.

Everything is a compromise.

--
 
Last edited:
IMO, the high pass filter at the I/P of an amp with GNFB is a must. Even the best trafos will saturate, when the amp attempts to reproduce infrasonic noise. Where you place F3 depends on how good the "iron" is. Approx. 17 Hz. is right, with decent magnetics. If the amp is less than 3 dB. down at 20 Hz. it IS flat enough. The lowest freq. a "standard" double bass makes is 41 Hz. Large drums and 32 foot organ stops are deeper, but how often do you play that sort of material?

In more than 1 way, the 6V6 is a scaled down, more linear, version of the 6L6. As the 6L6 works well in UL mode, I don't see trouble with 6V6s. Also, more than 1 "El Cheapo" builder has successfully implemented UL/triode mode switches.
 
Thanks for all of the input. The highest consensus reply I've read is that OPT is of crucial importance. It also seems that keeping the price tag under $250 will be a challenge. I will exclude from this price the soldering iron, punches, and metal drill bits.

Where the great variability lies is in the recommendation for amp to build. This does not surprise me, as everyone's tastes are different, plus many of you are the designers of such amplifiers. And with this, I vacillate between wanting to spend the fewest dollars possible to get a decent sounding amp to a no-compromises golden-tube-magic last-amp-you'll-ever-need build.
 
No, they are manually operated. Greenlee punches are the "Gold Standard" and are priced accordingly. This set is not as good, but, for the occasional builder, adequate.

Finally watched a video on how the knock-out punches work. Now I get it.

At the risk of sounding dumb, would my Dewalt Titanium Pilot Point Drill Bit set be adequate for metalwork, for anything small such as pilot holes for the punches? (I believe my set is this: DEWALT DW1361 Titanium Pilot Point Drill Bit Set, 21-Piece - Jobber Drill Bits - Amazon.com)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.