[zygadr]

Hi Guys,
Have built a breadboarded single channel version of this amp with the following changes:
1. no input transformer, replaced with cascade, direct coupled 6SN7 input stage, then cap coupled to the first ''driver'' 845.
2. no interstage transformer between 845's..........used a 0.22 uf copper foil in oil coupling cap.
3 using 5U4GB rectifier instead of 83



Everything else is as is..........sort of  .......a few less chokes...........no funds

Sound is fantastic! ............there is something magic happening here when a big ol' 845 drives another 845
I have built other amps before, including 300BSE's, but this amp is special.......no doubt about it

It's just a rough slap together at the moment, but any suggestions from the better informed are welcome

[JoshK]

Sakuma's approach is indeed interesting.  A hardline engineer would scoff at the bandwidth limiting step up transformers.   However, he makes it work for his application.

Without the interstages, its hardly a Sakuma design IMHO, however, you have borrowed one of the key elements;  a serious driver for a serious output tube.

I would try to rework for 3 stages max ( with 6sn7 cascade > 845 > 845, that is 4).  The easiest way to do this in my mind is to use a capable driver than has higher gain.  Think 211 or 801, maybe a 46.   211 would require higher B+ though, I think. 801 would work good at that voltage.   I am surprised the 845 even turns on at 450V (500v less 50v bias).

You can't argue with success.  If it works and sounds good, it is good, take my comments as just suggestions for exploration.

[zygadr]

Hey JoshK.............your moniker is freaking scary man!!................do you allways look like that?

Thanks For the driver suggestion. The cascaded 6sn7 was, and still is a proven winner in most 300b driver circuits, so I pinched that as I had plenty of the valves at hand and knew that it would have enough grunt to blast the first 845 driver tube as well as sufficient gain to use it without a preamp.

I did try a small cheap Hammond 124D interstage as a quick experiment in the initial construction phase and DID like the sound........but bass was non existant..........no surprise with that one

Problem is that,as you say, the less transformers the better and a ''GOOD'' interstage costs heaps in most cases.

On the plate voltage, believe me............at the 450 plate volts mark, the 845 is incredibly linear and good sounding

If you do a search on the audioasylum forum on ''LOW POWER 845 SE'' you will find many posts on those who have successfully built an 845SE at low voltage..............it can be done, and done well

[nullspace]

If you ask Mike @ Magnequest about low-power 845's, he'll go on and on and on about one a friend of his made using the StereoSound recommended operating points: 430V on the plate, -51V grid to cathode, and 62mA into a 5k load. Good for 4W.

John

[zygadr]

John, did he go on and on in a good way............positive?

[zygadr]

I'm still tweaking the pre stage 6sn7's to see what they're capable of, but so far so good..............brilliant in fact.

Sakuma's design is known for it's radical schematic, and the low value pot and input transformer, as admitted by Sakuma himself is an actual ''tone control'' to be used in conjunction with his equally controversial preamplifier designs.There is also mention of the fact that his designs have low gain.............therefore my use of the 6sn7 cascade.

All in all, it's an interesting philosophy............very culturally Japanese.........but fascinating just the same.

I have built a couple of variations of the low voltage 845 SE and each one sounds much better than it should considering the operating points in use.
It was when I used an 845 in front of another 845 that I realised how that driver's sound or lack of sound can contribute to the overall toanlity

 I have managed to obtain four Chinese standard 845's which for some reason are very hard to get at the moment, so I consider myself to be lucky in this case.

[nullspace]

John, did he go on and on in a good way............positive?

Yes; Mike's a big fan of that operating point. Whenever anyone brings up low-power 845s he suggests going with the StereoStereo operating point.

He also keeps throwing out, for a low-power 211: 480V plate, 40mA current, and -20V bias. Into a 10k primary, it'll squeak out 2800 milliwatts. I'll admit to having a set of parafeed output transformers and plate chokes to build that one, but haven't gotten around to it yet.

John

[JoshK]

There are two amp designs I've been kicking around in my head for my new projects.  I have to finish my speakers first before i work on any more amps though.

My system will be active and my "midrange" horns (~150-800hz) are suppose to be an honest 105db/w with the driver I've chosen, so I don't need a lot of watts.  For this bandwidth, I am contemplating a lower voltage GM70 amp inspired by Dowdy's schema.



This is a 3C24 into a GM70.  The 3C24 has a mu of 23.  The GM70 has a gain of 8-11 iirc, so higher than an 845.  I think Dowdy said this makes about 7 watts (I didn't do the math, but this is more than enough).  For the PSU, I plan to replicate John Swenson's PSU for his 813 transconductance amp.  I have all the parts on hand for that, so why not?  It happens to be perfect voltage and current for this amp. 

For the upper mid horn (800hz+) I am thinking of doing a 801A amp driven by a 112A (or 26) and doing stacked supplies like Jeff Davison has been so aptly demonstrating.

[nullspace]

Jim's GM70 amp certainly does get rave reviews everywhere it goes. The OMA guys were supposed to have it at RMAF in 2008, the one year I made it out there, but apparently there were some hiccups and it didn't make the journey. Perhaps the biggest disappointment of my trip was not getting to hear that amp.

But, not something I'm comfortable enough to build myself. ~500Vdc is my self-imposed limit. It's not that I'm particularly clumsy or absent-minded, but it would definitely scare the bejesus out of me to be working on something like that.

John

[JoshK]

My KT88 E-linear amp has voltage ~550VDC.   The thing with for me with working with these voltages is not touching anything when turning them on.  No probing.  Set up the measurements and then turn power on and when read.

[mgalusha]

I'm with Josh, no probing on really high voltage stuff. Clip leads connected before power on, stand back a bit and then power on. Seeing the meter hit 1050V makes it very clear that pre connecting stuff is the right choice. 

[zygadr]

GM70 amp looks very interesting............would like to know how it sounds

John, a low voltage 211?.............I didn't think that was possible. The guys over at the AUDIOASYLUM way back, never even considered it when the question was brought up.

[nullspace]

No reason why it wouldn't work. But, I don't know anyone who has built one and I haven't even heard of anyone doing it. Can't say I'm suprised, though -- it's a lot of work for a sub-3W amp.

John

[FullRangeMan]

HI,
Iam a GM70 fan.   According this very informative doc http://diyparadise.com/nikcgm70.html
by Mr.Nicholas Chua the GM70 work fine with low B+, only 450V, Grid -70mV and fixed Bias, output is only 8W for the GM70 last a million hours.
He inform ''Fixed bias gives a more dynamic sound too.''
Regards, Gustavo

[zygadr]

Reconnected the cheap Hammond 124D interstage last night intead of just a coupling  cap between the two 845's.

Wow!..........what a difference. This may be a crappy little interstage, but it's good enough to reveal the severe compromise in just using cap coupling between two seriously large tubes.

As I'm using a 500ohm, 50watt resistor in line before the transformer's B+ fly lead, I guess I'm taking a bit of the load of the trannie?............it barely gets warm even though it's rated at only a mere 5 watts!

Anyhow, the sound is pure dynamite - explosive dynamics, an effortless, powerful sound, crystal clear.

Lots of promise here

[JoshK]

As I'm using a 500ohm, 50watt resistor in line before the transformer's B+ fly lead, I guess I'm taking a bit of the load of the trannie?............it barely gets warm even though it's rated at only a mere 5 watts!

I am not sure if I understand what your saying, but according to Kirchoff's law, current is preserved.  So however much current is going through the resistor is also going through the ITx.  You are likely knocking down the B+ though quite a lot by dropping voltage across the 500ohm resistor.  This changes the op.pt and maybe reduces current as a result.  It all depends on how you biased it up.

[smbrown]

Zygadr, interesting post, thanks. I may have missed it, but what are you using for an output transformer? Also, what 845's in use? I've looked at getting some to experiment with, but price that I've found seems kind of steep. Thanks!

[zygadr]

Josh, the 500r 50w resistor gets quite hot and yes,.......I used it to drop the plate voltage down a tad. If I connected B+ directly to the interstage, IT became hot ..........now it's only just warm.

The circuit is biased up at the moment (with the interstage in place) at around 54 ma on the 845 driver and 64ma on the output 845. So, it seems that everything is working o.k, but I can definately use only a single 6SN7 stage instead of the cascade...........a bit TOO MUCH gain

Hi smbrown, I'm using LANCROFT 5K output transformers built by Bob Sugden many years ago here in Australia. They are fantastic transformers! ............compared very favourably to TANGO transformers in a shoot out once
The 845's are the bog standard Chinese cheap one's. I have 4, and know what you mean by the price of these things going through the roof!

Just a question to the experts out there...............the Hammond 124D??????.............the specs indicate that it's ''minimum'' frequency responce is 150 - 15k at -1db at 1khz, 5watts max.
Does this mean that it's bandwidth is really lousy?..............I ask this because I connected the breadboarded mono prototype to my large two way horn system and was surprised at how good it sounded

The mids and highs were fine but I'm not sure about the bass............it was there and made my CD stand ''buzz'' occasionally...........but I can't say that it was noticeably ''full'' or maybe it was just my perception/expectation?

 Perhaps the 150z spec had me convinced that there would be poor bass performance
After playing for 2 hours, there was no noticeable sound dererioration(saturation ??)..............weird?..or not?

I'm not suggesting that this is the interstage bargain of the century, but I presumed that sound quality would be severely compromised by it's use.................I was wrong.

I guess that you get what you pay for, but lashing out big dollars on some of the Lundahls that don't state input and output impedance specifications has me hesitant at the moment.

[zygadr]

Is it just me that's ''jagged'' the effective use of an el cheapo interstage, or am I not being treated seriously

Surely someone out there has had either a good or bad experience wiith the HAMMOND 124D?

[JoshK]

I don't know many who've tried it.