6550/KT88 SET amp aka "Uber Beam Machine"

jrebman · « **Reply #180 on:** 24 Mar 2008, 09:43 am »

Jeff,

I think with a few component changes and values, you could drop a 6h30 in the simpleSE and have all those same tubes. I also don't see any reason why you couldn't just build one half of the SimpleSE board (or copy the circuit) and save yourself a lot of money. Not to say or imply that Mike is overcharging for those amps, I'm sure they are nice (wouldn't mind having a set of mid monos or V4s myself), but we are DYI'ing here

.

If you want a tube rectified 8 watt monoblock using 6h30s and a high damping factor, check out Jim Hagerman's Cymbal amps. I've heard them many times and they are very sweet amps.

-- Jim

jeffdavison · « **Reply #181 on:** 24 Mar 2008, 03:30 pm »

No way on earth would I buy those amps...I'm sure his price is fair, but I vowed to never buy a commercial amplifier. I have built my assembly skills quite high. Just that my design skills are a bit lacking. When I saw the Quickies layout and components, it was exactly what I wanted to BUILD. I wouldn't know j.s. about swapping capacitor ad resistor values to sub one tube for another.

Perhaps you could tease me with the values you mentioned that would be needing change in the Simple SE to use the 6H30?

Thanks and regards

JD

JoshK · « **Reply #182 on:** 24 Mar 2008, 04:46 pm »

Jim is right that it would be relatively easy and quite aligned with George's philosophy (higher gm tubes are favored) to sub the 6h30 for the 12at7. However, lower gain/input sensitivity is to be expected. If you are using a tube preamp with substainative gain then this might work well, but passives will have a difficult time obtaining enough gain.

I am not sure where the 6h30 likes to operate. Some googling might turn up some operating points said to work well for the 6h30 (plate voltage and/or B+, current and bias voltage) from other diy'ers. Or you can look to see if you can find triode curves for the 6h30. I know Broskie published some way back when. We can help you work out the necessary changes if you will do the leg work to obtain the necc. data. Sound fair?

jeffdavison · « **Reply #183 on:** 24 Mar 2008, 04:57 pm »

As I mentioned... even if I find operating points for the 6H30... I would have absolutely no clue how to use the data to determine what parts would need to be changed for whatever values.

JD

DaveC113 · « **Reply #184 on:** 24 Mar 2008, 05:50 pm »

You'll only need to change 2 resistors, one at the anode to set CCS current... this can be selected from a chart on the CCS datasheet. Then, use ohm's law (v=i*r) to select a plate voltage using the cathode resistor. I am going to use a JJ ecc99 for the driver tube. Generally, you select the op point based on a load line drawn on the triode curves, but theres so much range I think its better to use an op point that has proven to give good results for the application. I just did some searches on diyaudio to see what op points people have had good luck with for the ecc99.

There are a few sites that offer free downloads of the 1952 radiotron book, thats a good thing to have around, it explains tube theory and design pretty well.

Dave

JoshK · « **Reply #185 on:** 24 Mar 2008, 05:56 pm »

Quote from: jeffdavison on 24 Mar 2008, 04:57 pm

As I mentioned... even if I find operating points for the 6H30... I would have absolutely no clue how to use the data to determine what parts would need to be changed for whatever values.

JD

But that is what I said I would help you out with.

I quote myself

Quote

We can help you work out the necessary changes if you will do the leg work to obtain the necc. data.

jeffdavison · « **Reply #186 on:** 25 Mar 2008, 07:40 pm »

All kinds of 6h30 data:

http://klausmobile.narod.ru/testerfiles/6n30p.htm
http://www.tubes.ru/techinfo/HiFiAudio/6n30pdr.html
http://www.glass-ware.com/Vacuum_Tube_Live_Curves.html

JD

Quote from: JoshK on 24 Mar 2008, 05:56 pm

Quote from: jeffdavison on 24 Mar 2008, 04:57 pm
As I mentioned... even if I find operating points for the 6H30... I would have absolutely no clue how to use the data to determine what parts would need to be changed for whatever values.

JD

But that is what I said I would help you out with.

I quote myself

Quote
We can help you work out the necessary changes if you will do the leg work to obtain the necc. data.

JoshK · « **Reply #187 on:** 25 Mar 2008, 09:23 pm »

So a good place to start is 150V on the plate, 20mA of current and grid bias of -7.9V. This equates to a 395 ohm cathode resistor. Then you need to set the CCS to 20mA, assuming it can take that much dissipation.

jeffdavison · « **Reply #188 on:** 29 Mar 2008, 05:02 am »

Here's an interesting design using a different sort of front end tube. An EF806S or an EF 804S (Different pin out) is known as an extremely quite tube used for Nueman microphones and is a sub for the EF86. First time I've seen it in 6550 / KT88 SE amp.

http://www.tubebuilder.com/images/schematics/singleended/kt88_se2a.gif

JD

JoshK · « **Reply #189 on:** 29 Mar 2008, 07:07 pm »

Interesting, but I wouldn't suspect the EF86 to be able to drive the KT88. There was a recent discussion about an EF86 driving an EL84 (super easy to drive) and the consensus was it might barely do it. There is another schematic I've seen whether they use an EF86 in pentode mode, like this, in an aikido adaption front end to drive a 300Boid. The logic is that the EF86 has tons of gain and sounds really nice but it doesn't have the guts to drive the 300b directly, so the Aikido's cathode follower buffers the EF86 and provides the low impedance output and current drive to drive the output. It also just happens to do noise cancelling for the EF86. Pentodes have extremely poor PSRR.

FWIW, the EF86 is a low noise pentode, but that is just relative to pentodes which are much noisier than triodes. It is just that the EF86 has a lot of gain, much more than most any triode. In a mic preamp you need a lot of gain. So the EF86 was developed so that high gain applications could have an option that wasn't too noisey. The Aikido is a nice choice here because it cancels out the power supply noise.

thread on diyaudio
Schematic

Freo-1 · « **Reply #190 on:** 29 Mar 2008, 07:39 pm »

FWIW,

I am using a pair of 5687's as a driver tube for a pair of 6550A/KT-88 outputs, and that tube works quite well in that application.

JoshK · « **Reply #191 on:** 29 Mar 2008, 07:48 pm »

I'm sure you're right. The 5687 should be able to drive the KT88 with aplomb. But you say you are using a pair to drive a pair. Do you mean PP, PSE or just the stereo channels? Also, since the 5687 has dual triodes are you parralleling the sections?

Freo-1 · « **Reply #192 on:** 29 Mar 2008, 08:01 pm »

Quote from: JoshK on 29 Mar 2008, 07:48 pm

I'm sure you're right. The 5687 should be able to drive the KT88 with aplomb. But you say you are using a pair to drive a pair. Do you mean PP, PSE or just the stereo channels? Also, since the 5687 has dual triodes are you parralleling the sections?

Sorry, should have stated it more clearly. It's a similar topology to the after market boards that use a pair of 12AU7 tubes to drive a Dyna MkIII. (One 5687 provides the gain, while the second acts as the phase splitter to a PP configuration).

It's a Citation II, with the 12BY7's replaced with 5687s.

IMHO, a worthy modification for the Citation II.

JoshK · « **Reply #193 on:** 31 Mar 2008, 03:55 pm »

Anyone else make any headway on theirs?

I received the aluminum sheet I am using for my top plate this week so I drilled out most of it this weekend. This time I took my wife's advice to glue down the graph paper rather than just tape it down and it is easier to work with this way. I used 3M 77 spray. This isn't quite as sticky as the 99 type, which I hope makes it easier to clean up after the fact.

I have some dark cherry wood that will work great for the base. I need to obtain an adaptor for my router in order to use the template guide for my dovetail jig. I have 3 routers, two of which are very popular and the template piece from a Sears industrial dovetail jig fits none of them.

I am trying to decide whether I want to send out the top plate to have it powder coated and painted when I am done punching holes in it, and then where to send it.

I've also thought about setting a copper single sheet on top of the aluminum and finishing the copper, experimenting with a patina. However I have two dilemna with that. First, I am not sure if aluminum and copper react when placed together for periods of time. Second, how to affix them together so they won't vibrate.

JoshK · « **Reply #194 on:** 31 Mar 2008, 04:00 pm »

Found my answer to the aluminum/copper mix... in leyman's terms its bad.

one commentary on mixing metals

I am not sure I will have an electrolyte present, but still its better to not risk it.

jrebman · « **Reply #195 on:** 31 Mar 2008, 04:06 pm »

Quote from: JoshK on 31 Mar 2008, 03:55 pm

Anyone else make any headway on theirs?

Well, I sent in the money for the transformers

Oh, and I bought some caps to try in the PS.

-- Jim

jrebman · « **Reply #196 on:** 3 Apr 2008, 07:52 pm »

Iron has arrived and now it's time to get on with the task of laying out the top panel. The first question that comes up is how close can the outputs be to each other? I never really thought much about this and honestly don't know how critical it is. These are big 9-10 pound chunks of iron (even the PT, which is the same core size as the outputs) and getting all this on a reasonable size plate with appropriate distance to the power transformer is going to be a matter of juggling fractions of an inch.

Anybody have any formulae or guidelines for OPT spacing?

One thing I really like about these is their utter simplicity -- 3 primary leads -- plate, B+ and UL tap, and two secondary leads (6 ohm and ground).

I picked up some Axon TrueCaps -- 75uF and 41 uF for gain and power tube cathode bypasses respectively. These are somewhat smaller, and a good bit less expensive than the Solens, and people say better sounding as well. We'll see. If not, it's always easy to go back and put some BGs, Silmics or Panasonic FMs in.

-- Jim

JoshK · « **Reply #197 on:** 3 Apr 2008, 08:11 pm »

The magnetic field interaction drops with the square of the distance, much like SPL. So the more distance the less it matters. I have no idea what absolute magnitudes we are dealing with. I just know that George shows a couple of his SimpleSE amps with the OPTs really really close, like < 1/2" apart. I am providing about 1.5" between my OPTs.

Reversing the orientation of one versus the other helps a lot too. Putting Power transformers and audio transformers at 90º to one another is a good idea too. I only know these simple rules, nothing beyond the cursory.

jrebman · « **Reply #198 on:** 3 Apr 2008, 08:21 pm »

Ok, well as I have things now, I'm somewhere between you and George, but I'm not going to turn one output -- the PT, yes.

I think this is going to turn out pretty well.

-- Jim

mgalusha · « **Reply #199 on:** 4 Apr 2008, 05:10 pm »

Quote from: JoshK on 31 Mar 2008, 04:00 pm

Found my answer to the aluminum/copper mix... in leyman's terms its bad.

one commentary on mixing metals

I am not sure I will have an electrolyte present, but still its better to not risk it.

Josh,

I did this a few years ago but the aluminum was anodized, so that likely insulates it enough to prevent any problems. I used Gorilla glue to bond the copper to the aluminum. I would use epoxy if I was going to do this again. I will say I have had zero problems with the copper on aluminum chassis. It's the preamp in this (old) photo:

.

It looks pretty nice in person. I coated it with some clear spray at the time but it was a little cold out when I did it and it dried a little funky but not enough to make me re finish it. I chose to put the screws through both layers and counter sink them but it would be smoother to countersink the lower plate and epoxy the screws so there were no holes and fasteners in the top plate.

mike