I did it!

for the DIY ST35, the Dynakit and every other PP EL84

Postby corndog71 » Thu Sep 27, 2012 11:49 am

WA4SWJ wrote:Sorry about the recent problem you had with the resistor. Shouldn't be too hard to resolve. Congrats on a great build!

One thing I tried that improved my ST-35 a lot was to replace the 12AX7 with a 5751 (lower gain tube) and the 12AU7 with a 12BH7. This made a big difference in my amp. I never went back and am still using them.

Good luck with the amp! What fun this hobby is.


Thanks for your compliment.

Did you adjust any part values in order to better accomodate the replacement tubes or leave them stock? I would think adjustments would be called for for the best performance.
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Postby WA4SWJ » Thu Sep 27, 2012 12:04 pm

Corndog,

You know, I did not make any parts changes to try to further optimize it for the new tubes. My build used Shannon's diytube board so it would have been slightly more trouble to pull it out and make some tweaks. The new tubes sounded so much better than the old tubes I just left it as it was. It still sounds great.

Back in the forum archives there should be a picture or two of it posted. I haven't looked to attach a link for you. I also ran some squares waves through it until just below clipping and posted those as well. If I recall I got about 21 watts out just before clipping - but then my memory might be foggy. This is a great amp design.
Ed Long
User avatar
WA4SWJ
KT88
 
Posts: 650
Joined: Mon Dec 27, 2004 8:39 pm
Location: Belleview, FL

Postby corndog71 » Thu Sep 27, 2012 1:32 pm

Sounds like a fun experiment. Been meaning to check out those tubes for other projects.

Even though I'm missing my amp, I've been wanting to rebuild the power supply anyway. Just wait until you see what I do next!
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Postby EWBrown » Thu Sep 27, 2012 8:04 pm

I also did the 5751 and 12BH7A "mod" a few years ago, it made a noticeable improvement with the DIY ST35 with the MQ565 OPTs [:)

I did not make any resistor value changes.

So I ran into a problem the other night. The amp was playing fine one minute and then I started hearing some intermittent pops of distortion and then I lost a channel. The R12 (50 Ohm 5W) resistor suddenly started glowing and smoking. I hadn't made any changes to the circuit. It had been running fine for over a year.



If it still measures 50 ohms, and you're getting 15VDC drop through it, then the B+ current draw is 300 mA, which is about 2X too high for the ST35. Check your EL84 / 6BQ5 idle currents - the easiest way to accomplish this with the EFB circuit, is to measure each EL84's plate voltage (on pin 7), they should be within a volt or two of each other, if one of them measures significantly lower than the others, then that tube is "hogging" too much current.

That burnt resistor looks to be rather small for a 5W power resistor, I'd use a 5W or even 10V "cement" or ceramic type WW power resistor, or better yet, a C354 or equivalent (1.5H, 200 mA, 56 ohms DCR) choke, if there is room inside of the chassis.



/ed B
Last edited by EWBrown on Mon Dec 03, 2012 8:06 pm, edited 1 time in total.
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby corndog71 » Sun Sep 30, 2012 2:57 pm

I really appreciate the helpful suggestions. Many thanks for that!

Ok, I found some replacements for the 50 Ohm 5W resistor and installed test points for pin 7 of each EL84.

So here's what I got.

Going into R12 - 310 VDC
Coming out of R12 - 289 VDC
B+ to the tubes - 260 VDC

Pin 7
Left channel
V3 - 268 VDC
V4 - 266 VDC

Right channel
V5 - 263 VDC
V6 - 274 VDC

Out of curiosity I check the EFB bias and found something even weirder:

Left channel - 1.09 -1.19 VDC

Right channel - .91 - .99 VDC

Those are the readings I was getting while turning the pot. Not even close to where it's supposed to be.

Amazingly the amp still played music! I know all of this needs to be addressed before I put it back into regular use but I just thought that was interesting.

So I guess I need a new set of power tubes. If there's anything else you guys can think of to measure or test let me know.

I bought some new power supply caps and I'm going to redo that whole section. I'll even pick up one of those chokes to replace R12.
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Postby dcgillespie » Mon Oct 01, 2012 6:41 am

Corn -- Something is very wrong at this point, and it does not sound like new output tubes will resolve the problem, either.

You are indicating that there is a 21 volt drop (310 - 289) across your new R12 (50 ohm) resistor. That equates to a current flow of 420 ma -- or 4X to normal amount. This is supported by your (basically) 1 volt reading across the bias test points in each channel. Such a reading equates to 200 ma per channel, or again, 400 ma of total current draw. I'm sure R12 did go out spectacularly! With the EFB modification, this resistor is normally dissipating just under a watt, so a 5 watt device should be more than ample for the job.

This problem is all consistent with no actual bias being provided to the output stage. Therefore, I would determine the cause before operating the amplifier any further. With EFB, this could be because:

1. The EFB circuit is no longer sensing the B+: i.e., the 360K resistor has opened, the bias pot is defective, etc.

2. Either or both caps in the EFB circuit has/have shorted,

3. The regulator itself is shorted, or

4. The old standby, wiring errors.

I have no doubt that the issue lies somewhere in these areas. The only components within the EFB circuit that have any significant dissipation associated with them at all are the 360K resistor, and the regulator. Therefore, I would assume that the issue lies with one of these two components (assuming that there are no wiring errors).

Assuming a 2 watt resistor was used (it normally dissipates .36 watt), as well as adequate heat sinking to properly dissipate the (typically) 1.5 watts that the regulator is dissipating under normal use, then no component failures have been noted with EFB (that I am aware of) when these cautions are observed. In my original EFB installation, the EFB regulator only becomes mildly warm after several hours of use when bolted to the rear of the SCA-35 chassis, and shown in my article. With your rather unique installation, if your regulator is regularly operating so hot that you cannot touch it for long at all, that is a sign that heat stress may have taken its toll on the regulator, and it finally gave up the ghost. If that is the case, you might want to consider an alternate method of heat sinking the regulator better.

I'm sorry you are having the problems you are, but I have no doubt that you will find the issue soon enough.

I hope this helps!

Dave
dcgillespie
KT88
 
Posts: 399
Joined: Sun Mar 08, 2009 5:34 am
Location: Ball Ground, GA

Postby corndog71 » Mon Oct 01, 2012 9:00 am

Thanks for your input Dave.

I was kinda thinking it might be either the bias pot or the regulator or both. The heat sink on the regulator never got real hot as far as I remember but it has been bumped several times. I can certainly try a different mounting arrangement. The pot defintely seems defective.

Good thing I bought multiples of each part.

It looks like it'll be a few weeks before I can get it rebuilt.

In the mean time my Class D Audio SDS-258 is filling in nicely.
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Postby corndog71 » Mon Oct 01, 2012 11:10 am

I have another question for the experts.

As I wrote before I wanted to redo the power supply anyway and have placed an order for some different caps. I'm currently using Panasonic TS electrolytics 47uF, 120uF, 47uF, 120uF.

I noticed the original ST35 uses 60uF, 40uF, and 20uF.

The values I was going to order were 45uF, 110uF, 45uF, 110uF. But sadly the 45uF caps are out of stock. Would a 33uf cap suffice? I'm thinking it should be ok but wanted to double check with you guys.

I'm also wondering if I really need a 4th RC since the original ST35 only uses 3. I could save a few bucks by skipping the 4th cap if it's not necessary. If this is the case I would end up using 33uF, 110uF, 110uF. I think I read somewhere that the first cap should be smaller and have seen this implemented in other amps.

I suppose this might also require a change in the associated resistors. I'm currently following the Audioregenesis SCA35 circuit.
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Re: I did it!

Postby corndog71 » Mon Dec 03, 2012 1:02 am

Update!!!

The power supply was not working. The pot was clearly malfunctioning. I had spare parts to start over with but also wanted to try something new. I decided to try the new Clarity Cap TC Series 700V film caps! Designed just for this application. I bypassed the larger 110uF caps with ESA .1uF caps. I'll add some more details later.

http://i576.photobucket.com/albums/ss20 ... G_6924.jpg

http://i576.photobucket.com/albums/ss20 ... G_6923.jpg

Holy crap this thing sounds amazing! Talk about veils being lifted, this thing sounds like one of those fancy amps I'll never be able to afford! I had a feeling it was going to sound good but the actual results have left me utterly awestruck! Even at low volumes the music seemed effortless and expansive. This is a seriously major tweak that I highly recommend!
Last edited by corndog71 on Wed Jul 12, 2017 3:41 pm, edited 1 time in total.
some people call me Rob
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Re: I did it!

Postby corndog71 » Thu Jan 03, 2013 7:43 pm

After a week or so of installing the Clarity Caps I realized my Matched quad of Valve Art EL84 tubes were no longer matching which I'm sure was caused by my parts malfunction. So with a new payday I splurged and got a matched quad of Gold Lion EL84's. I've been a big fan of the inexpensive Valve Art tubes for several years now and while I got better resolution from the Gold Lions the Valve Arts are not far behind.

Since I was adding all new filter caps and redesigning the power supply around them I wanted to make sure I didn't have another pot failure. Looking back at Dave's original paper on the EFB mod I was reminded that he used a full size pot in his experiments. I had had great success using PEC pots for my preamps over the years and so I got one for the EFB this time around. Seems to work well.

It's been a month now and I had an audiophile buddy of mine come over and listen for hmself. He almost didn't believe what he was hearing. He said it didn't sound like 17-18 watts but twice that. The dynamics startled him! Particularly when we listened to classical music. I threw on my 180g pressing of Kind of Blue which he's heard hundreds of times and was very familiar with. He shook his head and said it was the best he'd ever heard it. There was no lag or apparent distortion. He mentioned how there was no longer a slight haze to the sound confirming what I heard in the changes. Really good recordings were crystal clear with the kind of timing and rhythm you usually hear from the likes of BAT, Linn, and McIntosh. My friend was also impressed by the bass. Bass guitars and standing bass instruments were easily discernable in the mix. Actually every instrument stood out if you focused on it and could be followed easily. He was both disgusted (in a good way ;) ) and thoroughly impressed by my system performance!

Sources: Rega P25 with Shure V15VxMR cartridge into a Cambridge Audio 640P phono preamp. Pure i20 iPod dock with digital coax going to a Musical Fidelity V-DAC MkII. I have either AIFF or Apple Lossless files on my iPad.

Preamp is my Ultimate Quickie which is a fully modded version of the Bottlehead Quickie you saw at the start of this thread. Battery powered.
http://i576.photobucket.com/albums/ss20 ... 5e3986.jpg

Suffice to say I'm pretty darn happy with this stereo! Now I think I might try building a pair of Ultimate Mark III's. *)
Last edited by corndog71 on Wed Jul 12, 2017 3:41 pm, edited 1 time in total.
some people call me Rob
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Re: I did it!

Postby corndog71 » Fri Feb 08, 2013 8:40 pm

Whoa! This thread is getting a lot of views. If anyone has any questions or wants to know more about my build feel free to ask.
some people call me Rob
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Re: I did it!

Postby MLVA123 » Tue Mar 19, 2013 9:28 pm

I'm one of the lurkers.
I've looked at this thread several times - it's really inspiring. I'd like to try it, but I'm not at your expertise level so I guess I would need to figure out where to start!
Mike
Tube Noob
MLVA123
 
Posts: 33
Joined: Fri May 16, 2003 6:11 pm
Location: Richmond, Va

Re: I did it!

Postby corndog71 » Wed Mar 20, 2013 8:14 pm

I appreciate that, Mike. I don't really consider myself an expert. Believe it or not, I'm still learning. I've been tinkering with tubes on and off for a little over 15 years. I think I learned the most building Dynaco and Bottlehead gear. I've rebuilt an ST70, built a brand new pair of Mk IV's, and with this ST35 really wanted to challenge myself. I probably spent a month just going over the schematics from the original Dynaco design, the DIY Tube changes, and the EFB mod. I would redraw the schematics in order to better learn them. I would then draw the actual layout so I could see which parts go where. I double, triple, quadruple checked my drawings with the original schematics. I made lots of mistakes on paper in order to minimize mistakes with the real parts. Trust me when I say I was very intimidated by this project. I almost didn't do it. I almost gave up several times and just bought Shannon's board. (Sorry, Shannon! I really appreciate your efforts!)

In the end I really just wanted to do it my way. I wanted the EFB circuit. I didn't want individual tube bias controls. (more complicated!) I wanted a box to contain it all. And I really wanted the simplicity of point-to-point wiring. The best part of all of this is that I now have full confidence that I could build any of the Dynaco amps from scratch in any layout style I want. There are many similarities between them and they really are very simple circuits. Before building this amp I built a Class D Audio SDS-258 - 250 watts with 97% efficiency. It sounded better than I expected and drove all of my speakers impressively but still seemed lacking in resolution compared to my Dynaco amps.

Your amp looks great! If I were to build another one I would definitely build it like yours with the tubes and trannys on top. Interesting color choice too. Do you have a shot of the interior?
some people call me Rob
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

Re: I did it!

Postby MLVA123 » Thu Mar 21, 2013 5:35 am

That's cool, corndog. I do much the same with my crossover designs. I learn a lot from studying the schematics, and redrawing them. Lately, I've been getting into using Jeff Bagby's PCD - it's a wonderful tool for crossover modeling. Once I get the frequency and impedance response I'm looking for, then I can breadboard the circuits and play around. I need to step up to the more complex amplifier circuits, there is so much more to learn here! Speaking of circuit simulation, I recently started using CircuitLab - it's a browser-based circuit design and simulation tool, and it's free. There are tons of public designs and templates available if you are looking for something to start with.

As far as my old ST35 goes, here's a pic that I sent to Shannon way back when I was troubleshooting my initial build. There's a bug in the hook up - if you want a quick challenge, see if you can spot it. haha

Image
Mike
Tube Noob
MLVA123
 
Posts: 33
Joined: Fri May 16, 2003 6:11 pm
Location: Richmond, Va

Re: I did it!

Postby corndog71 » Thu Mar 21, 2013 11:58 am

I think you forgot to hook up the flux capacitor. ;) That board looks different than the ones pictured in the C and D documents. I see 8 filter caps. Thanks for the links. That circuitlab one looks interesting. I still have all of my notes so I can shoot you a copy of my drawing of the driver board if you're interested. Separating the power supply from the driver circuitry really helped me out. With my layout it's still a bit of a maze but I find it much easier to digest. The power supply is really simple. I also incorporated a star ground using a pair of solid core 16AWG wires. It's all coming back to me now. Fun stuff!
some people call me Rob
corndog71
 
Posts: 105
Joined: Thu Feb 03, 2011 10:45 am
Location: midwest

PreviousNext

Return to stereo 35

Who is online

Users browsing this forum: No registered users and 4 guests