• 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.

Need 3-5W tube amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I believe 5W would be enough for the project. But if that will cause any changes in the schematics and will require any tuning/experimentation I can go with 10W or more to avoid that as my experience in this area is zero. Even tubers' terminology is like a foreign language to me :)
 
rpi,
the difference between 5W and 10W is 3 dB. It's just one "click" on volume pot.
If 5W is not enough for you, you should go to a minimum of 20W.

5 W. is the yield from a SE "12" W. pentode. The power handling variations among O/P "iron" models affect the low frequency handling ability. The Edcor GXSE transformers are specified down to 40 Hz. The 15 W. model allows for going deeper (open loop) when fed only 5 W. and provides magnetic headroom to deal with the deep bass error correction signal. O/P "iron" core saturation sounds BAD.

BTW, it takes 50 W. to produce twice the perceived loudness of 5 W.

I believe 5W would be enough for the project. But if that will cause any changes in the schematics and will require any tuning/experimentation I can go with 10W or more to avoid that as my experience in this area is zero. Even tubers' terminology is like a foreign language to me :)

5 W. can be PLENTY, with the appropriate speakers. Paul Joppa provides us with a useful rule of thumb. Joppa's Rule states that in an "average" size listening space, an amp/speaker combo should be capable of producing 102 dB. SPL peaks at a 1 meter distance. So, 96 dB. or better sensitive speakers are in order. This under $300 DIY setup will be good enough to drive costly Klipsch Cornwall speakers. :D DIYing a suitable alternative will cost a heck of a lot less $.
 
Lingwendil, I mentioned already that I have a pair of speakers which are similar to the ones which you referenced:
Amazon.com: DROK15W Mini 3'' HiFi Full Range Speaker
Here I'm using them with HiFiBerry Amp2 amplifier:

cd-002.jpg


As expected they sound pretty flat without box. So I will try them first in the box and see if they work well. If not I'll either try your proposal or 4" size.
 
The spec. sheet for the Faital driver "Lingwendil" suggested contains (IMO) a very interesting number. Notice the Qts of 0.7, which suggests excellent performance in a bass reflex setup. FWIW, I did, and still own, a 4 inch high Qts driver in a 0.55 ft.3 enclosure whose ports tuned the setup to 55 Hz. The simulation was flat within 3 dB. down to 50 Hz.

While that sort of cabinet volume is not available in a "boombox", a well thought out bass reflex setup should get the 3 dB. down point lower than the documented free air resonance of 110 Hz. Port the enclosure to the rear.

To be fair, that DROK driver's Qts of 6.7 is also attractive. What I see as an important negative is the 4 Ω nominal impedance combined with the 88.5 dB. claimed sensitivity. As 2.83 V. into 4 Ω is 2 W., not 1 W., the "true" sensitivity is 85.5 dB. :mad: Unlike an output transformerless (OTL) SS amp, a tube amp's O/P power doesn't increase into a 4 Ω load.
 
Here is the frame assembled which still needs a lot of adjustments - make threads invisible etc.But it gives the rough idea about available space. The central enclosure will be approximately 6"x6"x8" - WxHxD

frame.jpg



Here is the sketch again just for reference:


sp.jpg
 
Last edited:
A problem with UL mode "finals" is the substantial CMiller the small signal circuitry has to drive. Fact, the 'X7 triode is a wimp whose drive capability is poor. Full pentode mode finals clearly avoid its limitations. The gas discharge screen grid regulator maximizes full pentode mode open loop linearity and also provides a cosmetic plus. Full pentode mode "finals" are not crap. Several manufacturers, including Fisher & Scott, achieved excellent results.

I agree if the driver is limited to the 12AX7, however other tubes like the 6DJ8, 6N23P, 6N1P, 12AY7 and many others have no issues. I'm sure there are plenty of driver tubes capable of handling C-Miller with distributed loading feedback.
 
Suggest you have a look at some of the Baxandall tube amplifiers. His amplifier designs would work well and there is one around the power you indicated. The 3-3 works but power is very limited. I remember building one when about age 13 (60 years ago!) but moved on to other amplifiers within a very few years.
 
I decided to try a simple approach first - Spud Amplifier with 6E5P Tube Hopefully I can reuse the most expensive stuff - transformers if I'll decide to go different route. Here are the transformers which I'm going to buy: Output transformers - GXSE15-3.5K (2): EDCOR - GXSE15-3.5K Power Supply transformer - XPWR058 (1): EDCOR - XPWR058

I hope that output transformer should work fine for both cases: Spud and Magnavox. Please let me know if the power transformer will also work for both?

Also I need a wattage for resistors in the Spud amp. It would be very helpful if anybody could provide this info. Thank you!
 
No, a single O/P transformer will not do for both a 6E5P "spud" and a "knockoff" of Magnavox's circuitry.

The Thai member clearly stated that he has high efficiency speakers. Using the low efficiency DROK speaker in combination with that "spud" amp will have you listening with your ear next to the "boombox". :bawling: IMO, better to forget the whole thing than throw money away on something obviously unsatisfactory.

Use the correct voltages and the correct impedances, for any given O/P tube type. TANSTAAFL always holds true.

A fairly decent compromise for handling both 8 Ω & 4 Ω speakers is purchasing "iron" with a 6 Ω secondary. You will have trouble into 4 Ω nominal, if the impedance curve exhibits dips, especially in the power hungry deep bass region.
 
Which speaker parameter does tell you that this is efficient speaker or not?

Eli Duttman, have you repaired your scanner? ;)

I still have a hard time to understand the final schematics which you are trying to describe across multiple pages in this thread. Where to place that OA2 tube etc.
 
Last edited:
Member
Joined 2015
Paid Member
The Edcor GXSE15-3.5K output transformer is perfect for the spud amplifer, but it will not be a good choice for the Magnavox with pentode mode EL84 tubes. The GXSE15-5K will be better suited for this duty. If you buy the GXSE15-3.5K and the spud amplifier will prove to be weak for your speakers, your best alternate option would be an amplifer with EL34 output tubes, not the Magnavox schematic. With a properly sized power supply, a EL34 tube paired with the GXSE15-3.5K will deliver more power than the EL84. EL34 power tubes and a beefier power transformer will cost slightly more, of course.

I believe that GXSE15-5K will be good enough for the 6e5p. Power will be reduced slightly but the amplifier will still work satisfactorily. If your backup plan is the Magnavox, then buy the GXSE15-5K.

Speaker impedence is important on tube amplifiers. You must take a decision if you want the best result. I believe that 4 ohms is the best choice for small and cheap speakers, because it is the most common value on computer and soundbar applications - your Drok drivers are probably something targeted at that market. I tried a few drivers like yours and also a bookshelf speaker specified at the same sensitivity. With the volume almost at maximum, the 6e5p sounds good at my normal listening volume. I like it more than my EL34 SE amplifier. But that's it. If I want more volume, this amplifier/speaker combination will not deliver.

On my 6e5p spud amplifier, all resistors are 1W with the exception of the 1500 ohm power supply dropping resistors. They are 7W, but 3W would have been enough. I use a transformer with 200V 100mA output winding and it remains stone cold. Edcor XPWR058 is almost twice as big and will be a massive overkill for this amplifier, but the HT winding is only 180V and therefore you will need to lower the value of the dropping resistors if you want to build my schematic. Actually 180V is barely enough for Magnavox, 200 or 220 v were more common for that schematic, to get 300V b+. If you want to build the spud and have the Magnavox as backup plan, it is best to select a Edcor with higher HT voltage, and then drop the voltage on the spud amp power supply by using a inductive input filter or a beefier dropping resistor. You will probably spend less money with a 1:1 cheap insulation transformer such as the one Eli Duttman listed on post 6: use a full bridge rectifier for the spud, and the voltage doubler for the Magnavox. You may forget the old style power transformer alltogheter if you want, and use a switched mode power supply instead. My personal preference is for the oldest style power supply with vacuum tube rectifier and no semiconductors at all, I will eventually rebuild the 6e5p amplifier this way. If you select this route, you need a power transformer with center tapped HT winding and maybe an additional 5V 2A filament winding for the rectifier tube. It will be expensive.
 
Last edited:
I said that this model would be good with "12" W. O/P tubes. It has a 5 Kohm primary and can be ordered with any of 4 secondary impedances.

Which speaker parameter does tell you that this is efficient speaker or not?

Sensitivity is (perhaps) a better word than efficiency. Forget the "hype" found on Amazon and look at the real provided info. Sensitivity (SPL) is stated as 88.5, which as previously explained has to be derated to 85.5, given the 4 Ω nominal impedance. So, the driver will produce a sound pressure level (SPL) of 85.5 dB. at a 1 Meter distance, when fed 1 W. of power. Remember Joppa's 102 SPL Rule and an amp of more than 60 WPC is in order. Each 3 dB. increase in SPL requires a doubling of the driving power. Now the 91 dB. sensitivity of the Faital 4" driver is not high, but it is a heck of lot better than that of the DROK driver. "Only" 16 WPC are needed to satisfy Joppa's 102 dB. Rule.
 

Attachments

  • DROK Speaker Info..JPG
    DROK Speaker Info..JPG
    101.7 KB · Views: 365
102dB SPL peak per speaker 1 meter is a pretty strict requirement for home use. Two equidistant speakers add essentially non-correllated, or +3dB. So 105dB SPL, two speakers 1 meter, allows for 20dB peaks above a mastering room's nominal 85dB SPL "0VU", which is very loud. Distance and dispersion eat up some of this - it's very situation and room dependent - maybe -6dB or -10dB ballpark.

But modern pre-squashed recordings don't use anywhere near 20dB of headroom, and 85dB SPL average is really loud, so there's a lot of extra room at the top. In the context of this project Joppa's guideline seems 20dB or even 30dB too stringent. JMHO, of course.

All good fortune,
Chris
 
Last edited:
Chris, it's a rule, not a law. Listening tastes, room size/condition, and L_RD knows what else enter the calculus.

I don't listen to that compressed stuff and very much appreciate adequate headroom. The thought of "squashing" a 100+ player Liszt or Suppe "potboiler" does not appeal to me.

I think it's quite fair to say that combining a 5 WPC amp with 85.5 dB. sensitive speakers is most unlikely to work out well.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.