• 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.

805 MonoBlock SE Amp Kit?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have been lurking for a while. As a recovering audiophile from the past, I am convinced that I want to do a monoblock based upon the 805 to restart my interest. Have been designing and mfg all sorts of electronics for several decades, so experience I have plenty of.

With contacts and suppliers in Taiwan, HK, China, & the US, for just about everything, I have no real issue of putting it together other than if I should and of what design / features.

The production side of me sez, if I'm going to do 2x units, why not do 30x or so to start. The design that I am leaning to is based upon Bauanleitung fuer einen 805-Verstaerker, von Ronald Koridon - really like the concept - but in monoblocks. Thinking of canning the B+ silicone in favor of valve rectifier.

The reason for the thread is to get some idea if there is any interest in the future kit. No, I don't know the price yet - as it is only in planning, but I am sure it will not be cheap - but will be reasonable. The iron is very pricey as you guys know. My guess it might be 4-6 months until ready to ship.

Right now I am working with some well known and not so well known trany mfg to get some ideas. It is really interesting, one of my 'normal' domestic suppliers used to source the Acrosound back in the day ...

Open to all suggestions and patter - perhaps it will get all of us that are interested in this style of amp an easier and better way to put it together - or on the other hand, it might kill the project!
 
I have done it in monoblocks . The 805 is a really good sounding tube, I have three NOS american makes RCA ,United , and less known CEI . Difficult tasks in this Nobu Shishido schematic is to replace the rare and costly WE437A and achieve a clean 3.25A 10v supply and solve hum. I was obliged to buy the very effective GuidoTent heater supply make. Replace the IITC interstage specially wounded by Tango, it is nearly impossible. I started with a russian driver the 6H45pi sounds good but it oscilates easily and my task for now is to change it for the Shishido's 12AU7 and or 12AT7 version , or maybe 6SL7. I think the idea of a kit is a good one but i am sure will be costly aimed for diehard tubesound lovers .

Regards
tt

An externally hosted image should be here but it was not working when we last tested it.
 
Hi mpuman,
Perhaps this project may be too much to chew on.I don't mean
to be sarcastic but like you say putting our heads together can make
for an achievable one.

1. Shishido's circuit may put off the pure single ended guys because it has
feedback from an extra coil at the output.It has an interstage and
high power rating of the output transformer which make it expensive.

2. It is operating the 805 at positive grid voltage vs conventional
single ended.The plus point is more power to drive inefficient high end
speakers.

3. There is another implimentation by Hirokuni Michimori at the 2008
European triode festival which he called Regulator Drive which does not
use interstage ( a saving) but still using a feedback coil at output.
This is more feasible at lower cost but still you need custom wind output.

ETF2008 Michimori

Single-ended triode Amps using zero-bias transmitting tubes

Check out the 838 circuit. at low anode B+ so easy on component choice.

So it boils down to standard single ended or much more power with
positive grid and feedback.

Then there is the question of anode cap on top or new one below socket.Will
you implement both for flexibility?

I know I can't afford this so count me out,hope this helps.If you really want to
do it it can be done but what of sound quality?.Shishido died in 1998.Maybe
Michimori can help with the transformer.

TugaTweaker,
Do you have gridstopper resistor 50-220 ohm at the grid
of 6C45? that should stop oscillations?

If you use tube regulator then only one type 5R4 is feasible,I know there are russian tubes but you want people to have easy replacement and unfortunately 5R4 is NOS so no current production tubes.
Singa
 
Last edited:
Keep the cathode feedback on the output transformer - this is the best sounding type of feedback for zero/positive bias transmitting triodes. Get rid of the interstage transformer and use a FET follower and bipolar power supply with a CCS in the -B supply. (aka Tubelab's Power Drive circuit) Try to DC couple the first and second amplification stages and then cap couple to the FET. Don't even think about using the crappy 6C45Pi. There is no reason why you can't use current production 6SN7/12AU7/6SL7/12AT7/6V6 etc. and get good results. I would stay with solid state rectification, you don't want a tube rectifier choking the dynamics of your amplifier.
 
Hi guys,

Thanks for the input - it all helps with my decision.

TugaTweaker, I too was thinking of replacing the WE437A with a 6C45, but JLH's comments (and others) kind of led me away from that and looking at some of the alternatives that are suggested. Do you have any of Shishido's 12AU7, 12AT7, 6SL7 version designs? I have heard that if you drive the 6C45 harder >10ma, it starts to be very nice.

I too was considering a Constant Current Heater supply for the 805 for the same reasons that Tent observed - I think it is the way to go - you have confirmed it. What do you think of DC or CC supplies for the drivers? I was figuring on DC.

Since you have built the 805, please explain further: "Replace the IITC interstage specially wounded by Tango, it is nearly impossible" - yes, ex-Tango is gone, but someone, somewhere must have the real specs. As I found, basically is a 1:.08 ratio around 2K5:2K as I recall - OK, 'special winding, etc. Seems to me a very wide band toroid might be a solution - but then again ... BTW, your amp pix looks very nice!

Singa, I looked at Michimori's designs and like some of his ideas, but I have issues with cap coupling ... it opens up all the tonal issues of it. Also, they are much more complex than Shishido's. I don't know, but seem to me simpler is better. An old friend of mine (almost around when tubes were invented) when I mentioned what I was thinking about, told me he did an OTL amp based upon regulators ...

As for the feedback, one of the things that Koridon to the design was lower it. Also could go the way of Cary in the 805 series by putting a pot /w switch to control the FB. That way even the pure Class A guys might be happier.

JLH, OK, you probably are correct about the HV supply, that was one of the things I was juggling with. I will stay with the diodes. The FET follower is a sound technical solution to the interstage, but then too, more complex, cap coupling, and more parts.

As I said before, the big cost is the iron - what I am trying to do is lower the cost of the iron by quantity and design. I want to stick pretty much to a 'classic style'.

Well, a conversation has started - perhaps others will join in and offer their 2¢ worth. Open to other SE designs, in the 30-50 watt range.
 
Mpuman ,thanks for finding my amp nice . I will post here the Shishido's 12au7 alone and 12AU7&12at7 switchable schematic versions , this one delivers 50W rms. The interstage Tango, now ISO, was sold to me by the nice people at EIFL Japan. It is wounded in special way that cancels DC at secondary winding, provided by the IITC design. Explanation can be found at product description of Wavac805 amp at their website . The feedback at 805 tube extends bandwith i think and I like it .To me , besides power the most important goal of this amp is that there are no caps at all in signal way , and that can be heard !!! And we know how much we need to pay for the nice teflon caps or for other best caps !
Regards
tt
 
Last edited:
Thanks TugaTweaker for the link. I really like Shishido's 12AU7/12AT7 version.

Too bad that the links to MJ articles are images and not actual text, as I do not read Japanese and the translators need characters. Have you ever taken a look at: MJ????? - ????? - neat, but I surely do not like the feedback cap nor the 22w. I will have to search more of MJ ... sure is a lot of good looking stuff for us.

As I do not have Shishido's papers, do you or anyone else have his notes on the IITC design? The descriptions that I see are rather vague and not geared for an engineer or something that I can spec a part with.

I am making some headway on the transformers.
 
I too use google translate - works pretty good.

What about C Core & R Core transformers? Seem to be from a technical side a no-brainer as they are inherently gapped - but from the subjective side - how do they sound? Much lighter in weight and cost? Has anyone actually tried them & listened side by side? If they sound the same, sure would lighten the load both in weight and cost!
 
I too use google translate - works pretty good.

What about C Core & R Core transformers? Seem to be from a technical side a no-brainer as they are inherently gapped - but from the subjective side - how do they sound? Much lighter in weight and cost? Has anyone actually tried them & listened side by side? If they sound the same, sure would lighten the load both in weight and cost!

Hi mpuman,
From anectodal evidence and what I've read C and R core
transformers sound good or at least you will get the attention of
audiophiles.:D Audionote UK uses C cores in their top notch products like
interstage and power transformers.So if you can use them why not?
Lundal makes it's transformers using C/R cores.Maybe they can custom wind for you?
Regards.
 
mpuman

A couple of thoughts about the 805 kit.

I built a pair of Moondogs many years ago and was very happy with the outcome.

Recently I decided to replace them with a 45 based amp, and finding nothing I liked in kit form I just built my own.

Now that I have done it I believe that I did the correct thing as I see that I had the opportunity to make the choices of what to use for the circuit, chassis, resistors, capacitors and transformers.

That is not to say that your choices will not be good but by building my own I made my own cost / sonic choices.

The additional work of deciding on parts finding them was a fun filled treasure hunt, and the mechanical work was not that difficult.

I have often since thought about another project and the 805 has received a lot of consideration but with Lowrthers I feel that they would be a bit much for my system.
 
mpuman
The additional work of deciding on parts finding them was a fun filled treasure hunt, and the mechanical work was not that difficult.

I have often since thought about another project and the 805 has received a lot of consideration but with Lowrthers I feel that they would be a bit much for my system.

Yeah! just kind of a little overkill! Might just fire the drivers into the next room!
Back in the 80's and early 90's I had audiopile low efficiency speakers and built all kinds of 800 series transmitter tube amps, including 805, 813, 808, ...and even 833's. Many were A2 of course. Fun stuff.
They were space heaters that made music!
I still have some knarly deadly power supplies.

But then I went back to high efficiency speakers and am now running SE 45 amps and the like. 3 watts will fill my 900sq ft room to nose bleed levels.
I still occasionally play with the big boys, just for the heck of it though.

BTW: Mpuman, if you are doing the amp for yourself then by all means use 866 rectifiers. They look cool and by absolutely no means do they adversely affect dynamics/transients, despite gross generalizations to the contrary. Better yet, use 866's for the output stage and 816's for the input/driver. However, for a commercial kit, silicon diodes might be a better choice.
 
I am investigating C & R cores. Seems to me a lot of voodo in iron! How much is for real and how much is illusion is difficult to judge. So much of it is subjective / smoke & mirrors. The illusion that more $$ is better. I really don't think that is always the case. As I have proved time & time again with other products, the key is in design & purchasing power.

My domestic magnetics guy is not really into audio - his contention is its 'mostly all **' ... then again he does not see why anyone would mess with tubes and 'hears' nothing wrong with mp3's - so he is not much help!

I got some quotes for some 'well named' transformer mfgs for the project. For EI nickel core version vs 'normal' HiB it's like $1400 vs $900 for each side! Is nickel worth $500 more (just for interstage / output)? Anyone with real A/B experience - perhaps in blind tests?

The idea behind my attempt is to make available a very reasonable priced package that is superior to the China 805 stuff and less than say Wavac. Obviously, the China 805 stuff is not using these style xfmrs!

Since it would be a kit, each builder could modify stuff like caps / resistors / tubes, but ideally the kit would be the best it could be for the price - including one's time.

The hope is to supply a complete kit in the <$1K a side range that is comparable to the $5K a side factory built units. From what I see now, we can do it for <$2K with the 'normal' HiB xfmrs. I want to get to the <$1K - then I think we will have something.

Next item on my list is the chassis. I have plenty of CNC machine shops that can do the whole deal as machined AL. I have CNC sheet metal shops that can do it in stainless steel. And lastly we could so an 'exotic wood' & metal combo. All have various advantages & disadvantages & costs. Any ideas?

The quest continues. I have always been into electronics. Having your hobby also as your vocation is very dangerous! Please keep the comments coming.
 
I am investigating C & R cores. Seems to me a lot of voodo in iron! How much is for real and how much is illusion is difficult to judge. So much of it is subjective / smoke & mirrors. The illusion that more $$ is better. I really don't think that is always the case. As I have proved time & time again with other products, the key is in design & purchasing power.

Well, there is a lot of art in transformer design. You are optimizing a large number of parameters simultaneously. What you pay for is experience.
Not that there isn't also a good deal of hype, but that is the norm for audio, sigh.

I got some quotes for some 'well named' transformer mfgs for the project. For EI nickel core version vs 'normal' HiB it's like $1400 vs $900 for each side! Is nickel worth $500 more (just for interstage / output)? Anyone with real A/B experience - perhaps in blind tests?

I have done one comparison. It was an output transformer, from a well known audio supplier. I compared a pure M6 (as I recall) to one with nickel stripes. It was not completely apples to apples because the striped version had a lower Idc spec and was parafeed configured. However, the difference was noticeable. I used the striped version.

I have compared all nickel input transformers to conventional ones that were very similar but not exactly the same. I drew the same conclusion.

Not exactly proper scientific comparison, but an anecdotal data point. It was enough to convince me to use nickel, when I could.

The hope is to supply a complete kit in the <$1K a side range that is comparable to the $5K a side factory built units. From what I see now, we can do it for <$2K with the 'normal' HiB xfmrs. I want to get to the <$1K - then I think we will have something.

That sounds like a reasonable price point. I doubt you would have any room left for your own profit though. That sounds right about rock bottom for the BOM.

Next item on my list is the chassis. I have plenty of CNC machine shops that can do the whole deal as machined AL. I have CNC sheet metal shops that can do it in stainless steel. And lastly we could so an 'exotic wood' & metal combo. All have various advantages & disadvantages & costs. Any ideas?

Gee, any of the above sounds nice, but of course we would have to see the concept to make a decision on materials.

The quest continues. I have always been into electronics. Having your hobby also as your vocation is very dangerous! Please keep the comments coming.

Yeah, ditto here. But my day job is in RF IC design so the two are so far apart, it almost doesn't have overlap.
 
hey funkytek, sounds like we have a lot in common. We have done some RF board level stuff here. RF ... talk about black magic, only audio is close in that department - you are closer than you think. But with digital RF, a lot of the art seems to have gone away (but I'm sure not on your side).

Not really in to this for much profit. Yes, we have to make something, but not that much. With our buying power, we still would have a nice but not excessive margin.

OK, more on xfmrs. Why not do them like they used to be done? That is paper bobbins instead of nylon with paper layered between layers? That's what they did back in the day. Perhaps the low tech way is still the way to go - but with better steel. My old time xfmr mfg sez it's the way to do them. I am willing to give them a try - after all they have been making audio xfmrs for 60+ years. Perhaps I will also have them do a couple with nickel just for the fun of it.

I also have a guy working on a more exotic design with some very special core materials that he contends are very promising. Will see.

Lastly, I was planning on CC for the heater supply of the 805. What about CLC? Looks like 22,000uF X 25mH X 22,000uF would give plenty of filtering while providing kind of a 'passive' CC. Just seems to me that when one puts a gain stage in the circuit (typical of CC regulator design ala LM338 etc), stuff is going to happen to effect the audio. Or am I going too far?

Separate filament and B+ xfmrs will allow easy sequenced power up and standby modes. Even more iron! Looks like 5 or 6 pieces of iron, 3 valves, some resistors and caps. Gee, it's like something out of the 1920's (less the regulated bias supply for the 805).

Well, with some good ideas floating around, looks like the next step is to get some iron wound up and get a couple test beds set up so I and others can take a listen. Have a couple guys in the Pittsburgh area that are 'audiophiles' to help with the subjective side. Others are welcome when we get that far. If you are close by, let me know.

Actually once I get some protos wound, if there are folks out there that actually have Shishido 805 designs, I would be very happy to ship them some xfmrs to audition vs their ex-Tango, etal iron ...

I guess the reason I even did the posting is to give me the motivation to turn the crank. So far it has helped.

OK, I've got a plan. Time to put together some monster passives and stuff to have parts to work with. Other than SMPS and motor controls, have not done much power and HV work for a while - need parts. If nothing else I'll waste some $$, but get back to some hobby fun!
 
Lastly, I was planning on CC for the heater supply of the 805. What about CLC? Looks like 22,000uF X 25mH X 22,000uF would give plenty of filtering while providing kind of a 'passive' CC. Just seems to me that when one puts a gain stage in the circuit (typical of CC regulator design ala LM338 etc), stuff is going to happen to effect the audio. Or am I going too far?

Switching power supplies are your friend. I modified some 5v switchers to 6.3 to power an 811A. All decent switchers have isolated secondaries so you don't have to worry about grounding issues. Run two 5v supplies in series. One thing to keep in mind is that any silicon you use will need to be properly heatsinked.

My opinion is that filament chokes are impractical at high currents. A two dollar linear regulator makes much more sense.


Separate filament and B+ xfmrs will allow easy sequenced power up and standby modes. Even more iron! Looks like 5 or 6 pieces of iron, 3 valves, some resistors and caps. Gee, it's like something out of the 1920's (less the regulated bias supply for the 805).

Well, with some good ideas floating around, looks like the next step is to get some iron wound up and get a couple test beds set up so I and others can take a listen. Have a couple guys in the Pittsburgh area that are 'audiophiles' to help with the subjective side. Others are welcome when we get that far. If you are close by, let me know.

Actually once I get some protos wound, if there are folks out there that actually have Shishido 805 designs, I would be very happy to ship them some xfmrs to audition vs their ex-Tango, etal iron ...

I guess the reason I even did the posting is to give me the motivation to turn the crank. So far it has helped.

OK, I've got a plan. Time to put together some monster passives and stuff to have parts to work with. Other than SMPS and motor controls, have not done much power and HV work for a while - need parts. If nothing else I'll waste some $$, but get back to some hobby fun!


Hey I'm in the Pittsburgh area as well :)
 
The hope is to supply a complete kit in the <$1K a side range that is comparable to the $5K a side factory built units. From what I see now, we can do it for <$2K with the 'normal' HiB xfmrs. I want to get to the <$1K - then I think we will have something.

I am working on a Simple 845 SE stereo unit (but could be configured as mono-blocks) @ 20 Watts RMS per channel. I should be able to post the schematics and BOM by the end of January. No kit, but just a basic design to try and get more DIYer's into transmitting tube amps.

I think I can get the design BOM to around $750USD if you are frugal in buying the parts. This will by no means be designed with the "best" parts, but just a starting point for people to try. They can experiment and try different heater circuits, front-end circuits later, if needed.

The designed will be built and verified, not just an idea.
 
Hi THD+N - sounds like your 845 could be a good entry point. I want to do the 805 for more power since a lot of PP guys need more power for their inefficient spkrs sys. This will give them an easy way to give a try to SE.

The kit idea is geared towards diy'er naturally - I think that we will offer semi-kits as
well (e.g. less glass, less chassis, etc). That way the true DIY guy can get the main parts and roll some of his own. I'm guessing that a lot of guys will want the full kit - assuming that the sound review is VG - which I expect it will - otherwise we don't have a product.

Spoke with yet another domestic iron guy that has been around for 60+ years. I think we may have a good fit on the iron for performance vs cost objectives. I think we are real close to that <$1K a side (depending upon options). I'll know more as the week progresses and time allows.

astouffer, I have a lot of experience with designing switchers. Although their noise can be outside of the audio range, way too much of a chance for a production kit. I have determined that CCS for the 805 will be what we will do. If you get a chance, drop me a PM and let talk - I'm sure that we can be an asset to each other.

In the next week or two, I'll post our proposed sch for everyone's comments - good & bad. Then the wait for the various samples of iron ...
 
Hi THD+N - sounds like your 845 could be a good entry point. I want to do the 805 for more power since a lot of PP guys need more power for their inefficient spkrs sys. This will give them an easy way to give a try to SE.

Why not just 'breathe' on a commercial Chinese 805 amp ? That would solve your prototype issue at a stroke . At the end of the day 40 watts from an 805 is only 3dB more than 20 watts from an 845 . 20 watts will hit 100dB peaks easily with 88dB speakers . Unless the user has a huge rooom , far more than they'll ever need .

astouffer, I have a lot of experience with designing switchers. Although their noise can be outside of the audio range, way too much of a chance for a production kit. I have determined that CCS for the 805 will be what we will do. If you get a chance, drop me a PM and let talk - I'm sure that we can be an asset to each other.

With a cap input supply it can be done but the CCS will need a large heatsink and even a ring of two will dissipate several watts .

I personally think that an 805 kit is not worth it , especially with the complexity , costly output transformers and high voltages . It will no doubt be a real money pit in terms of development and think of the support you'll have to offer when people's builds go wrong

Good luck !

316A
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.