HoSo56 amp project...design input sought

[CraigB]

What manufacturer 5879's do you guys have?  I've got a few Philips ECG and a couple RCAs, which look like they were all made in the same factory, or at least with the same materials.  Interesting to me is that each one of them sound a bit different from the other, more so than the difference between different 12AX7s, for example.

Anyone achieving success with other pentodes, like 6AK5 and 6AU6, etc.?

I'm hooked playing into a pentode.  Theres so much complex ring and zing and makes even the most difficult parts to play seem easier to do.

[tubenit]

Well, ....... I had one of my 5879 tubes go very very microphonic. Never heard anything like that before with a tube. I view it pretty much unuseable at this point. It was in my 56T combo.

So I am interested also in knowing what brands might be best. Mine was a Sylvania.

I'm also curious what "makes" a tube go microphonic. The tube started out just perfect. I didn't play the amp for about a wk & then when I fired it back up it had the microphonics. It seems like it was not a gradual change but a sudden shift to being microphonic.

BTW, my son was in town & he had a chance to play all three of my amps ........ and his favorite was clearly the 56T! He thought it was the most expressive. Second favorite amp to him was the "Blues amp" (schematic in SCH library).

With respect, Tubenit

[CraigB]

Found this explanation:

"Loose electrodes, the metal pieces inside a tube which conduct electricity, are affected by vibrations generated by the loudspeaker. This causes the space between the electrodes to vary, and as a result the electrical fields to vary within the tube. This changes sound into current vibrations, which is positively fed back through the tube, and an annoying howl results. These vibrations can affect the electrodes through the socket or the base, or by sound waves striking the glass envelope of the tube.

"To help control microphonics Fender surrounds some tubes with a spongy, sound-absorbing material. In other amps and old radios mounting the sockets on rubber may also help. If both of these precautions fail the tube is too microphonic to be used, and should be replaced. You don't necessarily have to throw it away, as it may work fine in an amplifier with a separate head and cabinet."

REFERENCE: Marcus, Abraham. "The Electron Tube: Service Notes" Radio Servicing: Theory and Practice, Third Edition, pg. 165, par. 2, Prentice Hall Inc, 1960.

One of my Philips 5879's was destroyed when I cranked the GA40 for the very first time  :'(  I've had the same thing happen with EF86.  I now make a shockmount subchassis for my preamp tubes, and actually on this build made a subchassis for the 4xEL84 too.  Works like a charm.  I will post a photo so you can see what I mean.

Glad to hear your son got to dig in and try the 56T and others.  Your combos are really cool with that nice woodworking.  I can see from the picture of the 56T how the soundwaves from the speaker would tend to pummel the 5879.

[tubenit]

OK, it's obviously a limited "test" but I compared the 5879 tubes I have.   2 Sylvanias, 2 RCA and a Raytheon.

Both RCA's sounded darker and perhaps smoother but had the least character/expression to them. The Sylvanias had the best chime, expression and harmonics, IMO ..........but one of them has gone microphonic in a relatively short period of time.  The Raytheon sounded reasonably close to the Sylvanias and also had nice harmonics but perhaps not as chimey sounding. I thought there was a pretty noticeable difference between the RCA's vs. Sylvania & Raytheon.

Not much of a review but I thought we could start comparing some observations about the 5879 since it's kind of a new tube to many of us.   Anybody else have a comparison or description they could share?

Tubenit

[PRR]

> I thought the 5Y3 should only handle a 20uF as a first filter cap.

The capacitance hardly matters. If you have enough uFd to hold ripple to usable level, you have BIG current surges.

What matters most is resistance. Some bottle-rectifier datasheets spell this out. A complete calculation is mind-bending. However a "right size" power transformer intended for tube rectifiers -probably- already has a reasonable amount of resistance. If it is a copy of a PT from a tube-rectifier amp, size suitable to the rectifier you want to use, you can probably use 16uFd or 60uFd. If it is like the lump I just found in my office, 2 Ampere rated output, even 2uFd will probably weld a 5Y3's guts together.

> One column reads 120V for both plate and screen and the next column reads 180V plate and 120V screen. Which way should I go there?

Neither. Look at the currents: 7mA! You almost never want that much current in an audio voltage amplifier. Reverb Drivers rarely suck that much. (However if you did need a gain of 100 with a 20K output impedance, and did not mind the brutal drain on your B+, it works.)

I lost track of what plate resistors we are using. But since audio preamp loads are in the general range 100K-500K, reasonable plate resistors are 50K-500K. Typical preamp B+ voltages are 200V-400V. Plates and resistors usually split the B+. So typical currents are 200V/50K= 8mA to 100V/500K= 0.2mA. 1mA-2mA is most common.

The 7mA suggestions are for choke loaded amplifiers. Chokes are very much better loads BUT at audio frequencies a "good" choke is not very good and very expensive. 6AK5 and kin are aimed at radio frequency stages, where a great choke is a few pennies.

So the Typical Operation conditions on the 6AK5 sheet are NOT directly useful for you.

And the difference shown for 120V versus 180V is very-very small. Most parameters change less than 5%, when real tubes vary 20% one to the next at the same conditions. It isn't clear why they show the two different voltages. Mostly, pentode plate voltage does not matter for small-signal work. But maybe the forms looked bare with just one set of conditions.

You will probably want MUCH lower G2 voltage. How much lower is hard to say.

> Shouldn't the screens be a fraction of the plates by like a ratio of 5:1 or is that plate/screen current?

The plate/screen current ratio WILL be around 4:1 for 6AK5 at any reasonable plate voltage. The screen is a picket fence with 80% open space. Throw BBs at it, how many go through and how many hit pickets? A high wind in your favor (high plate voltage) may funnel a few extra BBs between pickets; wind in your face (low plate voltage) and more BBs may veer into the eddies behind the pickets.... but as page 2 shows, the plate/screen current ratio is very consistent right down to where plate current collapses.

[bruno]

I read through the whole topic, and I think none of you guys tried a vox top boost tone stack, any particular reason why that is so? I was actually pretty much convinced that the 65 amps were using a variation of a top boost but with a ef86 as the first gain stage rather than a single triode.

[Geezer]

Hmmmmmm........sounds logical, since 65 pretty much says straight out that they have used Vox amps as a model for their amps!

Maybe I'll try it in the 7W version.

Or, you could build one & let us know how it works out! 

Gz'r

[bruno]

Hi Geezer, well everytime you see Dan Boul in his youtube videos talking about a new amp release he seems to have his book of vox by his side, and I've seen from other videos he is quite keen on vox amps... besides their first amps were pretty voxy inspired (soho included). I have a ac15 twin from music ground JMI, a sort of clone of a vintage ac15 twin, and I can tell for a fact that the london's ef86 channel does get similar tones, however that's a totally diferent front end from the soho.

As a matter of fact I'm just about to build a prototype with this type of front end just for kicks, however I will be using a brown vibrolux output/power transformer replacements, and make it usable for 4x6v6s or 2x6l6s (mostly 6L6s I guess), but if I can share what conclusions I come to I'll be happy to. I'll only start building after the 18th of september, as I'm on vacation right now. I'm guessing this will turn out, if everything goes right, much closer to their stone pony or monterey models, which is fine with me, since the stone pony has become a favourite of mine among their videos.

thanks

[Geezer]

the stone pony has become a favourite of mine among their videos.

I'll have to check that out......

[Baguette]

I read through the whole topic, and I think none of you guys tried a vox top boost tone stack

Isn't it the only difference between the Vox tone stack and the marshall (HoSo) stack the treble capacitor (50pF Vs. 500pF)???

[tubenit]

Hey Bruno,

I know you tend to like to design your own amps ........ however, it sounds like you're considering something similar to my
Carolina Blues Rocket (at some level). 

I have a SCH schematic and layout drawn right here:
http://www.el34world.com/Forum/index.php?topic=6368.0

I have not added the VVR yet, but plan to in the next few wks. I did add it to the 56T and it works fantastic, IMO!

Let us know how you like your build when it's done.

With respect, Tubenit

[bruno]

hey tubenit,

I do love trying to come up with my own things, but I also love trying out others designs, like your carolina blues special, with the dano centurian tone stack, I've tried it in a prototype chassis. Wonderful tweedy sounds, although the tonestack is a little different than what I was used to.

For this build I'm not considering anything special, I just want to try the front end of the 65 amps, and since the only set of transformers I have laying around the house is a brown vibrolux set, I guess they'll have to do. So yeah, the schematic you posted is pretty much what I'm after, although I'm going to install Dana Hall's VVR just like you did and like 65 amps uses. I'm going to scale only the inverter and the power section too I guess. I'm leaning towards trying the Vox tonestack though.

The differences between the marshall and the vox stack are not only the 47pf, if you check the schematic. However a cool ay to see how both react is the tonestack calculator.

thanks for the schematic tubenit, I have most of the components here which were spares from other builds. And I have a bag full of 6PS series caps, so I guess I'll use those. The only thing I need to buy are octal sockets. I even have a aluminum chassis ready.

PS: I'm not too sure why you guys use 33k to simulate the 68k resistors, in a single jack wiring, the 68ks purpose is to have the low input, if you don't use that input the 33k while emulating the traditional wiring, simply degrades sound, why not just leave it out? I've done so in several builds, one less part right?

[tubenit]

the 33k while emulating the traditional wiring, simply degrades sound

Not sure what you're saying here???    Paralleled 68k resistors are 34k which is close to 33k.

How does that degrade the sound?  It's what you hear plugging into many Fender amps.

With respect, Tubenit

[bruno]

well when fender used the two 68k it was in order to obtain the low input, if there is no low input, then no resistor is necessary at all, the only reason you guys are keeping it is to simulate the effect of having the 68k. A resistor in the signal path there simply degrades signal (which was a necessary bad thing in order to get that second input, but no in this case), and since the main function of the 68k resistors is to obtain the low input and there is no low input in a single input amp, well it just seems pointless to me to use any resistor at all.

[tubenit]

it just seems pointless to me to use any resistor at all.

Bruno,

At my home if I use a 10k input resistor ....... I sometimes pick up radio waves thru the amp. A 22k or larger eliminates that.

With respect, Tubenit

[bruno]

Hi tubenit,

well a shielded wire between the jack and the first stage grid will acomplish good isolation too... however in good truth I never had problems even without using the shielded wire and no resistor. But I guess it depends on the zone you live in.

I have even built an amp from a trainwreck express schematic, which has tons of gain, due to the way the amp was designed. No resistor between input and first gain stage's grid. No problem with radio interference. I even have a youtube clip of the amp being played really clean. But I've played it under all kinds of circunstances with no problems.

[sluckey]

Bruno, a point you may be missing is that series resistor (33K, 47K, 68K, whatever) and the interelectrode tube capacitance (grid to cathode) form a high frequency shunt filter. Hi freq parasitic oscillations or radio freq interference (RFI) that may be picked up by your guitar or amp input is shunted to ground through that hi-freq filter and the low impedance cathode circuit. This resistor is sometimes called a grid stopper or suppression rersistor. This could be important to you if you are playing in the vicinity of strong RFI. And including that resistor even on a single input amp will not degrade the guitar signal, unless there's another resistance involved that forms a voltage divider. Even then, I wouldn't call it 'degrading'. I'd just call it 'dividing'.   

[bruno]

Hi sluckey,

I'm well familiarized with grid stoppers... I just find them a bit pointless in a single input design unless you really are having problems with external interference. And then shielded wire helps there. A voltage divider is a different thing alltogether as your whole signal is dropped before hitting that first stage.

Just another idea, obviously, in testing we're basically free to do as we please... any way thanks guys!

[sluckey]

Well, you wanted to know why we use them. That's why. Plus, my brother-in-law works for Ohmite in the resistor painting section. He's in charge of the orange paint gun. 

[bruno]

Lots of external interference where you live then, hey? :P

So that must mean free resistors, lucky you, care to send me a few?

I'll tell you how the build went later on, might even try 7591 tubes much like the stone pony or 60s ampegs.

[tubenit]

Hey Bruno,

Enjoyed your video clip. Amp sounded good to me and very expressive. Liked your blues rifts and rhythms also. Good stuff. Thanks for sharing it.

Best regards, Tubenit

[bruno]

thanks tubenit, I'm actually a lousy player...

I built it, but the amp isn't mine, it's too loud for my purposes... besides it can do high gain stuff, this video showed the other side of the spectrum as I felt not enough videos had been done on trainwreck style amps clean tones. I usually stick with amps that have less gain, although I really like the trainwreck.

[sluckey]

Lots of external interference where you live then, hey? :P

Yep. This is the noisiest planet in the Sol system.

So that must mean free resistors, lucky you, care to send me a few?

All I can get are orange. Sure you want some?  

[bruno]

Lots of external interference where you live then, hey? :P

Yep. This is the noisiest planet in the Sol system.

well you never know, now do you?

So that must mean free resistors, lucky you, care to send me a few?

All I can get are orange. Sure you want some?  

I'm all stocked up to get me through this build, thanks!

[bruno]

hey dyna, I don't want to drift this topic away from the subject, so I can open up a new topic regarding the trainwreck copy I built with a couple of photos if you'd like.

However... I was playing at home, at very low low low volumes, and with a stratocaster with JM rolph 64 pickups, which are like a vintage strat and very low output. So yeah this thing makes chimy tone, it's a very dynamic and versatile amp, the most dynamic I've ever heard to be honest. It does get into very high gain territory.

please procede with the hoso topic...

[Dynaflow]

 No problem, I'll remove it, sorry for the hi-jack!

Regards,

Dyna

[arjepsen]

I've been scrutinizing the pictures linked earlier in the thread. I really want to try the circuit out with the ef86, as I already have that tube.
I found a few things that seem different to me, than the posted schematics.

To me it seems that the cathode resistor on the CF is not 100k. Comparing it to the 100k resistors to the right of it (the plate resistor for v2A and the "slope" (?) resistor for the tonestack), it seems to me that the color rings are: brown, grey, yellow, gold = 180k.

Now, I'm not too good at the theory here... what will the difference between 100k and 180k on the cathode do to the sound?

[fuzz]

Hi ,

I get myself registered here to thank you all for that great thread (and forum ) I've been reading for a while now .  I've just finished building an amp based on that topology (pentode/triode/CF/tonestack) with a SE 6V6 . Sounds huge ! the pentode has a great unsual sound . I started with a cap selector (as the "FAC" in old orange amps , and VOC ac15 I think) but I just took it off this evening , didn't bring something new to the sound.

I'm still working on it , devil is in the detail  , but I'll post a schematic once it's done !

I'm also very interested in 65 amps  designs ,  seem to be "original" designs (which is pretty rare in "boutique" market , they're all clones of something "vintage" ) . I"m trying to figure out what it's the "ventura" , the Tupelo/lil elvis could be also a great stuff to know 

fuzz

[Willabe]

I'm still working on it , devil is in the detail   , but I'll post a schematic once it's done !

Welcome fuzz and why wait?

Why not post it now and what your thinking and what your looking for? The guys love the process too as much as the end product.

You don't wanna steal away their fun, do ya????          

Start a new thread on it.


                     Brad         

[SILVERGUN]

I'm just a new kid here but I stumbled upon this one by accident and I'm really happy with the results...I shoved this Bogner Ecstasy overdrive circuit into a Sovtek Mig 50 in place of the pc board preamp and just left the power stage in tact and WOW....not too far off of that heavy overdrive sound near the end of this clip. Build it on it's own separate board and plug it in...If you don't love it,,,sell it to me (for parts cost only, of course )
Use this preamp, and you will not be sorry:

[jojokeo]

This can be fun but nothing really new there SG. Typcial cascading gain stages to CF design. I find it interestingly somewhat alarming, poorly designed, and excessive use of parts. Why put in high-end dumping techniques and then use high pass filters all at the same time? And why use voltage dividers which also affect tone and high end loss when split-load plate resistor set-ups could be used w/out these detrimental affects? But then who am I to criticize?

*BTW, how did this thread suddenly turn from EF86 amp project...design input sought to HoSo56 amp project...design input sought???

[SILVERGUN]

You know Keo, I was just trusting that Reinhold knew what he was doing...

I should've also checked the date and the direction of this thread before posting....chalk it up to a rookie mistake.
This is my first forum, and you guys have my full attention.....I can't get off the site because I'm learning something new every minute.
I got lucky with one build and I'm tryin to shove this preamp down everyone's throat

I'll get back to my shopping cart and try not to hijack any more threads

[jojokeo]

I'm not saying he doesn't know what he's doing, just that his circuit has contradictions going on and that I doubt Leo or Jim would've done the same if not for parts count alone. They were ingenious yet very frugal at the same time.

[fuzz]

I'm still working on it , devil is in the detail   , but I'll post a schematic once it's done !

Welcome fuzz and why wait?

Why not post it now and what your thinking and what your looking for? The guys love the process too as much as the end product.

You don't wanna steal away their fun, do ya????          

Start a new thread on it.


                     Brad         

yes why not after all !

I make it clear and post it !

[fuzz]

Hi guys,

I didn't forget about the schematic but I can't draw a clean scheamtic about this amps. I'll try to get a clean hand hand drawn one and post it (plus my camera is really annyoing with batteries) . Unless there's a website allowing drawing schematic ?

I'm also thinking about putting a switch to get the pentode alone into the cathodefollower + TS for a "clean"  and sending it into the 12AX7 for a kind of dirt side (or gain boost).

The EF86 (6N32P in my case) + 12ax7 is really crunchy even at low gain settings.

[John]

https://www.circuitlab.com/  Hi Fuzz, that is a web-based schematic program, free to use, you just have to register. Then there is also Express PCB which is a free download, easy to learn (even for me) and most everyone on here uses. Hope this helps!

[Fresh_Start]

https://www.circuitlab.com/  Hi Fuzz, that is a web-based schematic program, free to use, you just have to register. Then there is also Express PCB which is a free download, easy to learn (even for me) and most everyone on here uses. Hope this helps!

The best part about ExpressSCH besides being free is that tubenit already has SCH schematics for many amps.  Go check out the Schematics section here...

Chip

[fuzz]

https://www.circuitlab.com/  Hi Fuzz, that is a web-based schematic program, free to use, you just have to register. Then there is also Express PCB which is a free download, easy to learn (even for me) and most everyone on here uses. Hope this helps!

really cool webiste ! i tried to use "webschematic" but didn't get something clear as I wanted. But there're no tubes or my eyes are tired ?

I remeber that I  used to use expressSCH on an old machine , really easy software . I must have it on the hard drive .

I'll try to be quicker with that schematic !

I browsed few days ago and I was happy to see that there're a lot of good original projects in here !

[Isaac313]

I know this is old, but I'm curious if you ever got to this project and finished it. Also, It seems to me that the circuit would go more like this...

EF86-12ax7CF-tonestack-12ax7-PI-el84's.   The way I heard the amp, and peter stroud explain it, this is the way I would assume the bump circuit is done, again my assumption,  Bump circuit is they put a resistor underneath the mid resistor to make slight boost, and more mids and warmth, also disabling the rest of the tonestack from there.  Then using a relay to switch that resistor in, they use another pole of the relay or a second relay to connect in the tone control. Same with the level, relay that in, and it goes from there. I don't think they are actually switching between two different circuits in the tonestack section.

[marcoamf]

I'm interested in this topic, too, since I need a new project after I finish my first build (modded 5E3)!
This is an awesome work you guys have done. I wish I could find sound samples and/or videos on YT.

Apparently, the most complete schematic (leaving HoSo closer to the original) would be the HD VVR Draft #2, with the Mosfet based VVR (Dana  / Hall style) and the bump selector and level knob.
Rectifier tube should / could be EZ81 as the original (any other changes needed?).

I searched around but could not find additional comments on this version, nor a layout file that would be definitely handy.
In case we get a final version of this and layout, I can add a BOM file.

Thanks again to all contributors.

[marcoamf]

A recent video with Dan speaking about this amp.

http://www.ustream.tv/recorded/37928597

[tubenit]

Apparently, the most complete schematic (leaving HoSo closer to the original) would be the HD VVR Draft #2, with the Mosfet based VVR (Dana  / Hall style) and the bump selector and level knob.

I personally would pass on that one and build Geezer's version.  I'd use a 6AQ5 or 6V6 for my personal preference but the EL84 version is great also.

And I have built amps with both VVR and PPIMV that I liked.  However, on this amp, I'd use the PPIMV instead of the VVR.

With respect, Tubenit

[marcoamf]

Thanks, tubenit!
I'm not really familiar with PPIMV. Is it that 2.2M pot between the 12AX7 and EL84s? It's a dual gang pot?

Regarding the Bump circuit, do you think the one in the schem I sent is more complete?
I'm still noob in this, but it appears to me that it has Bump level and Bump tone dials (as in the original) and this schem you sent has two foot switches, one Boost (? middle?) and a Treble boost, as well as a Trim dial (volume?).

Thanks, again.

[tubenit]

It's a dual gang 250k pot with 2.2M resistor from wiper (middle terminal) to ground.

I'd use Geezer's bump/boost design instead.  You can always leave him a PM IF he doesn't chime in on this one.

It's Geezer's design and he has done ALOT of experimenting with it. 

With respect, Tubenit

[foghornleghorn]

Sorry if this has already been addressed, but Geezer's 3-18-09 "Final" schem shows a .1uF cap on the 5879 screen, but his layout shows 1.0uF. I'm assuming .1uF is correct as it's more or less in line with the RCA datasheet.

Just checking!

About to begin a HoSo56 variant build with a post-TS second 5879 driving a SE KT66 in Ghetto UL. Naturally I don't expect it to sound anything like Geezer's. Just need a break from building Dumbles, and this is using all major parts I already have, so no big $$ layout.

[tubenit]

.1 is correct.

About to begin a HoSo56 variant build with a post-TS second 5879 driving a SE KT66 in Ghetto UL. Naturally I don't expect it to sound anything like Geezer's. Just need a break from building Dumbles, and this is using all major parts I already have, so no big $$ layout.

Well that sounds very cool!  Can you please share the results when you get it done and let us know how you like it.

As an FYI,  I have found 5879's in an OD position can benefit in smoothness considerably by adding a cathode follower.  You could even use a mosfet CF in that manner.  (take a look at the D'Mars ODS as an example)

with respect, Tubenit

[foghornleghorn]

Thanx tubenit! I thought so.

I'll be happy to post it when it's done. Here's the current schem, I figure there's enough voltage out from the second 5879 to overcome impedance mismatch losses and still give the KT66 enough oomph to get 10 or 12W out of it.

If not, I bought some 6U8A to replace it and do the CF into the KT66.

The Ghetto UL I ran into on Ampgarage, it's a PP OT feeding the screen from the primary CT, then the plate. Not shown but I have a small 20H choke to put behind the KT66 screen gridstopper, to smooth out the screen supply and cut the voltage down to about 300V.

[foghornleghorn]

Thank you so much! I'm pretty much computer illiterate I'm afraid. I can use PSUD 2 and TubeCad and SE Cad. I will try to learn .sch though, it will take me some time I think.

V2 I changed to a 12DW7 because the 12AU7 makes a better CF. V2B cathode 39K or 47K. V3 screen cap is .05uF, to cathode not direct to ground. Cathode bias resistor is 1.1K (two 2.2K in parallel), bypass cap 4.5uF (nominal 3uF). This is set up to maximize voltage out, not gain. V1 is set up for max. gain. KT66 cathode resistor will probably start as 300r, as I'm told 250r is pushing the tube too hard (Shuguang KT66). 10uF nominal bypass caps I have measure 12uF. Voltage D in PS about 250V.

I've also been advised to try .033 TS caps in place of .022 because the SE output can use the extra lower end. Makes sense but I'll try it with .022 first.

Choke is Triad C7X 10H 90mA. Dropping resistors 8K2, 12K, 15K.

[tubenit]

FogHornLegHorn,

Thanks for the response.  I've edited it.  Any other edits needed at this point?  

The ExpressSCH is super super easy to use. Very intuitive.  It would take you all of 5 minutes max to use it given your skill sets with those other programs.  

Just click on a component and change the value when/where needed.

NOTE:

Not shown but I have a small 20H choke to put behind the KT66 screen gridstopper, to smooth out the screen supply and cut the voltage down to about 300V.

With respect, Tubenit

[tubenit]

FogHornLegHorn,

Are you on TAG?  And if so, what's your handle on that forum?

Just curious.

with respect, Tubenit