• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Kyoto EL84 Mono Block build

Jim Kiriakis

Active Member
Pursuant to TriodeLuvr's schematic based on Mullard 3 watt design principles utilizing Akai M-7 Terecorder power supply / output transformers attached here is a bill of materials as acquired and a rough draft layout (pins are wrong on the 6x4 rectifier). Poised to drill the chassis and begin.
BOM Kyoto.png layout.jpg
 
Register to hide this ad
Jim, the low frequency response of those particular transformers is very limited. Are you sure you wouldn't prefer to order a model with a wider bandwidth? The EF86-6V6 combination is very transparent and can do much better. Truth be told, even the Akai OPTs don't quite do justice to this combination, but they're well into the region of diminishing returns.

Jack
 
Jim, the low frequency response of those particular transformers is very limited. Are you sure you wouldn't prefer to order a model with a wider bandwidth? The EF86-6V6 combination is very transparent and can do much better. Truth be told, even the Akai OPTs don't quite do justice to this combination, but they're well into the region of diminishing returns.

Jack
Jack -

Good to know. I was not aware of that limitation for either said Edcors or the Akai M8 output transformers. May I prevail on you to suggest an alternative? Also please note I am using EL84 power tubes if that has any bearing on your recommendation.

thanks in advance,

Jim
 
You can build a very fine amplifier with the Akai transformers. When used with pentodes (as they were intended), the NFB necessary to tame IMD and provide reasonable dampening also flattens and extends the response. If you prefer not to salvage a pair from the tape decks, the Edcor GXSE10-5K appears to be a more reasonable substitute than their XSE variety. The CXSE25-5K is wider bandwidth still, but at considerably higher cost. The Hammond 1628SEA would also be excellent, and Electra-Print lists several suitable models in that same price range ($150 - $175 ea). Some will disagree, but I believe the benefit of these more costly transformers is likely to be difficult to hear in this application. Note that my advice is specific to the 6BQ5/6V6 pentode output stage using NFB. It would be different for a straight-through triode amplifier with triode output. One thing I want to be clear about - building the amp as shown in the schematic with those specific components does not in any way produce compromised results. I was extremely pleased when I first fired this up about 15 years ago. When I later also used it with Grado 32 ohm phones, I was astonished. It is totally engaging, with clarity and dynamics many people spend thousands more to achieve.

Jack
 
I was not aware of that limitation for either said Edcors or the Akai M8 output transformers. May I prevail on you to suggest an alternative?
If your plan is to only ever build one amp and you want it to "do it all" buying larger and more expensive transformers might be the best option.

But I've learned that you can't really assume that something larger and more expensive will necessarily sound better. And there are alternative strategies which allow you to have plenty of low bass without spending so much on OTs. This is especially handy if you expect to be building more amps in the future.

In general, the main difference between smaller, less expensive, OTs and larger, more expensive, ones is their ability to reproduce bass frequencies. The rolloff points will vary but the biggest difference is probably under 50Hz or so.

You might want to consider just how important this is to you. Everyone has different needs. YMMV, as always.

Some music has a lot of content under 50Hz and some has relatively little. Many speakers can't really reproduce much, if anything, below that either. And, of course, everyone's hearing is different too. These are all important factors. An online tone generator can be helpful in evaluating your needs.

I had my most recent build breadboarded for quite a while using a pair of little SE OTs pulled from an old console amp. I ordered a pair of Edcor 15w GSXEs for the build, which took about 3 months to arrive. The Edcors weigh in at 2lbs 9.5oz while the little console OTs are only 14.5oz. So I figured the bass would be better and they would be a step up.

This was the first time that I've purchased new, modern production, OTs for a stereo build. I've always used junkbox OTs scrounged from various sources. I decided to give the Edcors some break-in time - around 36 hours - without listening to them before giving them a serious listen.

The Edcors were really nice and, as expected, they did go lower in the bass.

But, for this amp, I actually preferred the little console OTs, which is not what I expected. They seem more extended on top and the mids are more detailed and liquid. They reproduce the subtle textures of strings and horns better.

A lot of the difference is in the soundstage. It seemed to be a bit wider with the little console OTs but they're especially better in terms of depth. The difference in detail is most evident in the way instruments seem to inhabit their own space. The effects of reverb on more modern recordings, or natural room sound on more traditional jazz, is much more evident. The Edcors reproduce a bit of that, but not nearly as much, and their soundstage is not 3D like the console pulls.

Don't get me wrong, the Edcors were impressive, and I plan on using them on another project. But when I A-B'd them I noticed the difference. The little console OTs produce more foot tapping and head bobbing, they just swing more.

Since I do occasionally listen to music that has more low bass, I picked up an inexpensive combination of an active crossover and a Class D amp. The total cost (~$180 shipped) was less than the cost of a single large, high quality, OT. Adding a sub is another option, of course.

I find the versatility of this approach very appealing. Since I know I'll be building more amps, I'll end up spending much less on each of them since my "bass system" can be used, whenever necessary, with any of them.

And, since it's harder for an amp and the main speakers to reproduce bass frequencies, there are additional advantages to shifting that burden to a system that's optimized for bass.

Such an approach may not appeal to everyone, of course, but it's something to consider if it meets your needs.
 
Wouldn't it be better to have C3 connected to the cathode of V1 instead of ground?

The voltages indicated in the schematic seem to be a bit off.

The total resistance in the cathode lead of V1 is 620 (R3) + 136 (R4 and R13 in parallel) = 756 Ohm. The cathode voltage is 1.1 V. So the cathode current would be 1.1 / 756 = 1.46 mA.

The anode current of V1 is (220 - 115) / 100K = 1.05 mA, while the screen grid current of V1 is (220 - 76) / 750K = 0.019 mA. The sum of these currents (= 1.24 mA) should be the same as the cathode current (= 1.46 mA), but they are not.

The current through R9 is 55 / 50K = 1.1 mA, while this current should be the same as the cathode current of V1 (= 1.46 mA), but they are not.

The current through R10 is 20 / 5K = 4 mA. So the screen grid current of V2 would be 4 - 1.1 = 2.9 mA at the most, and even lower if V1 passes more than 1.1 mA. But when looking at the datasheets for the 6V6 I think that at Vg2 = (275 - 13) = 262 V and Vg1 = - 13 V the screen grid current will be close to 4.5 mA.

The cathode current of V2 is 13 / 300 = 43.3 mA. But when looking at the datasheets for the 6V6 I would expect that at Vg2 = 262 V and Vg1 = -13 V, the cathode current would be close to 50 mA.
 
The EF86 screen bypass has traditionally been configured both ways. The same is true of grid 3. Feel free to experiment, but considering how successful this amplifier is in terms of my own listening experience, I wouldn't change either.

Voltage measurements were taken when I inspected the amp to recreate the schematic. I would expect small discrepancies for several reasons. First, I was pressed for time when I captured the schematic, and voltage measurements were not a priority. The amp wan't subjected to an extended warmup, so errors in I/E/R correlations are possible due to drift between measurements. Also, when measuring high impedance points where very little current flows, the DMM itself will skew the readings somewhat. Finally, the tubes in this amp are pulls from equipment built in the '60s. The outputs likely have thousands of hours on them now, and the preamps are not far behind. They don't respond like the factory curves. In the end, all the voltages on the schematic should be considered guidelines, not absolutes.

Jack
 
  • You can build a very fine amplifier with the Akai transformers. When used with pentodes (as they were intended), the NFB necessary to tame IMD and provide reasonable dampening also flattens and extends the response. If you prefer not to salvage a pair from the tape decks, the Edcor GXSE10-5K appears to be a more reasonable substitute than their XSE variety. The CXSE25-5K is wider bandwidth still, but at considerably higher cost. The Hammond 1628SEA would also be excellent, and Electra-Print lists several suitable models in that same price range ($150 - $175 ea). Some will disagree, but I believe the benefit of these more costly transformers is likely to be difficult to hear in this application. Note that my advice is specific to the 6BQ5/6V6 pentode output stage using NFB. It would be different for a straight-through triode amplifier with triode output. One thing I want to be clear about - building the amp as shown in the schematic with those specific components does not in any way produce compromised results. I was extremely pleased when I first fired this up about 15 years ago. When I later also used it with Grado 32 ohm phones, I was astonished. It is totally engaging, with clarity and dynamics many people spend thousands more to achieve.
Thanks Jack, completely helpful reply. The amp per your circuit now plays I am working through the QA process now. My immediate issue is a pronounced hum from both channels. Build facts differing from your amp - EL84, Hammond 369JX PTs, 6X4 vintage NOS black plate RCA rectifier tubes. Chokes are Hammond 155 and do deviate from stock Akai M8 which I recall to be 6H vs 5H for my Hammonds. I did not run PT simulation (yet) so it’s possible as a hum culprit which by ear sounds like 120hz. Your amp has floating drivers (ungrounded) and likewise I have yet to ground pins 2,7 and 8 on my NEC EF86s. My RH84, KG84 and Akai M7s are dead silent so I’d like to achieve that with your circuit as well. Next I will triple check my grounds and OT wiring. Also note I deleted the input volume control and headphone outlet.27254C22-8CA6-4606-8780-B45470146D01.jpeg
 

Attachments

  • 0BF817A9-CECA-4D72-838E-BCDD61011561.jpeg
    0BF817A9-CECA-4D72-838E-BCDD61011561.jpeg
    135.1 KB · Views: 48
If your plan is to only ever build one amp and you want it to "do it all" buying larger and more expensive transformers might be the best option.

But I've learned that you can't really assume that something larger and more expensive will necessarily sound better. And there are alternative strategies which allow you to have plenty of low bass without spending so much on OTs. This is especially handy if you expect to be building more amps in the future.

In general, the main difference between smaller, less expensive, OTs and larger, more expensive, ones is their ability to reproduce bass frequencies. The rolloff points will vary but the biggest difference is probably under 50Hz or so.

You might want to consider just how important this is to you. Everyone has different needs. YMMV, as always.

Some music has a lot of content under 50Hz and some has relatively little. Many speakers can't really reproduce much, if anything, below that either. And, of course, everyone's hearing is different too. These are all important factors. An online tone generator can be helpful in evaluating your needs.

I had my most recent build breadboarded for quite a while using a pair of little SE OTs pulled from an old console amp. I ordered a pair of Edcor 15w GSXEs for the build, which took about 3 months to arrive. The Edcors weigh in at 2lbs 9.5oz while the little console OTs are only 14.5oz. So I figured the bass would be better and they would be a step up.

This was the first time that I've purchased new, modern production, OTs for a stereo build. I've always used junkbox OTs scrounged from various sources. I decided to give the Edcors some break-in time - around 36 hours - without listening to them before giving them a serious listen.

The Edcors were really nice and, as expected, they did go lower in the bass.

But, for this amp, I actually preferred the little console OTs, which is not what I expected. They seem more extended on top and the mids are more detailed and liquid. They reproduce the subtle textures of strings and horns better.

A lot of the difference is in the soundstage. It seemed to be a bit wider with the little console OTs but they're especially better in terms of depth. The difference in detail is most evident in the way instruments seem to inhabit their own space. The effects of reverb on more modern recordings, or natural room sound on more traditional jazz, is much more evident. The Edcors reproduce a bit of that, but not nearly as much, and their soundstage is not 3D like the console pulls.

Don't get me wrong, the Edcors were impressive, and I plan on using them on another project. But when I A-B'd them I noticed the difference. The little console OTs produce more foot tapping and head bobbing, they just swing more.

Since I do occasionally listen to music that has more low bass, I picked up an inexpensive combination of an active crossover and a Class D amp. The total cost (~$180 shipped) was less than the cost of a single large, high quality, OT. Adding a sub is another option, of course.

I find the versatility of this approach very appealing. Since I know I'll be building more amps, I'll end up spending much less on each of them since my "bass system" can be used, whenever necessary, with any of them.

And, since it's harder for an amp and the main speakers to reproduce bass frequencies, there are additional advantages to shifting that burden to a system that's optimized for bass.

Such an approach may not appeal to everyone, of course, but it's something to consider if it meets your needs.
Read with appreciation FLaCharlie. I would certainly be interested in OT’s that provide the 3D soundstage your describe. Am optimistic about the Edcors. This build is an attempt to see what would result as I did use ElectraPrint in my recent RH and KG builds.
 
Looks like a nice build, sorry to hear about the noise. As I'm sure you know, hum can be caused by a number of things - ground loops, transformer coupling, PS ripple, etc. It can be tricky to troubleshoot, because hum thats induced by 60 Hz sources often contains harmonics. I would start by using clip leads to ground the grids of the 6V6s. That will take the preamps and the preamp grounding out of the picture. Also, if you have a scope, check the frequency of the hum.

Jack
 
I've been using ground busses lately. I've had fewer problems with hum doing so compared to Star grounds.
 
It looks like there isn't a center tap for the 6.3v heater winding? If not, a 100ohm or so resistor on each lead to ground, and removing the ground from that one leg may help reduce hum if you haven't done so already. That is if there isn't a center tap.
 
It looks like there isn't a center tap for the 6.3v heater winding? If not, a 100ohm or so resistor on each lead to ground, and removing the ground from that one leg may help reduce hum if you haven't done so already. That is if there isn't a center tap.
The center tap is not grounded. And I neglected to ground the heater leads per the schematic. Will try your suggestion(s) here and report back. Thanks!
 
Looks like a nice build, sorry to hear about the noise. As I'm sure you know, hum can be caused by a number of things - ground loops, transformer coupling, PS ripple, etc. It can be tricky to troubleshoot, because hum thats induced by 60 Hz sources often contains harmonics. I would start by using clip leads to ground the grids of the 6V6s. That will take the preamps and the preamp grounding out of the picture. Also, if you have a scope, check the frequency of the hum.

Jack
Jack -
Grounded the EF86 drivers' internal shield and grid (single wire ground to pins 7 & 8). Resolved the hum. Switched from my load / test speakers. Monoblocks sound excellent so far, plenty of bass through two way bookshelf speakers that aren't particularly efficient (86 db). Have yet to ground the heaters may or may not later depending on how it goes. Will try full range open baffle speakers next and report back to you.
 
That's good to hear. Those pins should be grounded, which is the reason they're shown that way in the schematic. The fact that mine doesn't hum with the shield accidentally floating is just luck.

Are you using the Akai output transformers after all, or did you order the Edcors?
 
Back
Top Bottom