DIY Audio Projects Forum

6EM7 power amp
Page 24 of 25

Author:  mwhouston [ 25 Nov 2013, 23:19 ]
Post subject:  Re: 6EM7 power amp

I decided to pimp Abstinence:

1 - Remove 22uf electros as first cap in PS and replace with 4uf\400V Russian PIO and snub with 0.1uf\630V poly
2 - Place to 100ohm\5W resistors between power tranni and rec. tube plates (control surge)
3 - Reduce last filtering\limiting R in PS to compensate for 100ohms on plates
4 - Replace orange drops with Russian Teflon 0.033uf\400V inter-stage caps
5 - Snub 100uf NP bypass caps (Ck) with 0.1uf polies.
6 - Replace the $24 5Y3 rec. tube with the $200 Sophia Electric mesh plate.
7 - Added sound deadening (bitumised Al) to bottom plate

I couldn't mount the 4uf cap direct to the bottom of the top plate so now is sits atop a Woodfords bourbon American Oak cork top. I had to reduce the HT to 260V and current to 88mA. Max mA for the Sophia is 90mA. I realise I am running it near max but I was not prepared to drop the HT anymore. Tube and caps are new so need some time. Sounds excellent though, lot clearer with much better sound stage. The giant Sophia is only a few mm from the power tranni but you can put the back of your finger on the tube. It is hot but not burning. Not worried.

Author:  mwhouston [ 26 Nov 2013, 16:25 ]
Post subject:  Re: 6EM7 power amp

I meant to include this image as part of the upgrade (pimping) I made to Abstinence. While I had the amp apart I added some bitumised heavy Al foil to the bottom plate for resonance and vibration control. Before I added the foil I held up the bottom plate by one of the brass cones and rapped it with my knuckle. It rang like an out-of-tune bell. I then applied the foil. The difference was huge. This time it just made a dull thud. No ringing and no vibration and barely any sound at all.

Usually I apply this stuff (comes in three sizes on a roll) before testing the enclosure, lid, bottom plate, top plate or whatever for ringing. I got quite a shock just how effective it is. It takes seconds to apply and is cheap when bought in bulk. You can see here also I add soft foam pads to the large heavy locking end for the brass cones. They came with the cones for other purposes but I like to use them here once more to add more dampening in the cone feet.

Author:  Suncalc [ 20 Dec 2013, 20:57 ]
Post subject:  Re: 6EM7 power amp

So, I had a snow day today and I've been cleaning up some amp designs to keep busy. To make a long story short, I've redesigned the circuit for the 6EM7/6EA7 amp. Of course, most people (especially Mark) will probably ask "Why?". The answer is that there were a couple of things that I didn't like about the original amp design. Not major problems mind you, just little things that kept nagging at the Engineer in me every time I looked at the design.

First, I didn't like how close the design point was to the maximum plate dissipation of section 2. The tube limit is 10W and the original design point was 9.8W. This left precious little margin if a power supply was a little over voltage or a cathode resistor a little low in value. So I've optimized the output stage design to reduce the plate dissipation to 9.0W (leaving a 10% margin) while retaining the output power and slightly improving the distortion characteristics. I also took the opportunity to improve the overdrive characteristics of the power stage to make it a little smoother in marginal overdrive conditions.

Second, I was just not happy with the gain and distortion characteristics of the driver stage. Ever since I designed it (September 2011), I've felt that the sensitivity of this amp was a little bit low; requiring a full 2.0v peak to attain full power. So I've redesigned the driver stage to improve gain and lower distortion. I've also included an option for a bypassed driver stage. In the new design, the amp requires only 1.72v peak to reach full power with the driver cathode unbypassed and only 875mV peak to achieve full power if the driver cathode is bypassed. The bypassed and unbypassed distortion numbers have also decreased from 2.5%/1.3% full swing to 1.6%/0.8% full swing. This change will make the amp sound a little cleaner.

So here is the new schematic:

The changes are not dramatic. The power stage has a new bias design with a little larger biasing resistor and a reduced bypass capacitor. Everything else about the power stage remains the same. In the driver stage, the biasing and load are different. It's only two (or three if you bypass the cathode) components but I think the changes really improve the design.

The only thing to watch with the new design is the Miller capacitance of the driver stage with a bypassed cathode. If you choose this option, be careful that the output impedance of whatever drives the amp is not so high that it begins to impact the high frequency response. With a CM of 194µµf the limit on the grid circuit total impedance presented to the grid is ≤41kΩ. With the 100kΩ potentiometer as shown, the max driver impedance is ≈64kΩ in this configuration.

As always, questions and or comments are more than welcome.

Author:  mwhouston [ 20 Dec 2013, 21:42 ]
Post subject:  Re: 6EM7 power amp

Due to the fact I was giving a demo of my 2-way inexpensive speakers to the Melbourne Audio Club a few night back, my system, was not setup the way I like. Though I enjoyed the 12" stereo subbies being driven by Abstinence (independent 200hz passive low pass filter) , the best setup is the subbies driven from their own 50W class D amp and 52hz electronic Xover and Abstinence only concerned with the 2-ways.

After the big changes I made to Abstinence just prior to preparing for the demo, I didn't get a lot on time to listen to It with all it's upgrade parts including the Sophia Mesh Plate rec. tube. System is back to normal now and I'm just starting to appreciate what an incredibly clear and gorgeous sounding amp this really is. Vocal tracks I have played in the past and know well, take on a whole new life. The upgrades are definitely worth while. Where I would never tell Matt how to design a tube amp or even what layout should look like or wiring method should be used I will suggest some components I think have really stepped an already good amp up one full notch.

If you can afford it the Sophia Mesh Plate 274B (about $220 shipped), 4uf Russian PIO\400V first cap in the PS, NP caps for bypass snubbed with 0.1uf polies, 0.033uf Russian Teflon inter-stagecaps, 20H choke, shock mount power tranni and choke and lastly fine wire-wrap wire for signal connection from RCAs to driver triod. I believe these components have resulted in a dead quiet amp (average 1mVRMS of noise) and it's almost surreal clarity. Look forward to you build Matt.

Two days back it was 40C here. Today a lot cooler.

Author:  Isophon [ 20 Dec 2013, 22:10 ]
Post subject:  Re: 6EM7 power amp

Hi Matt,

Thanks for the new schematics. I have a bunch of 13FD7 (very similar) so I might give it a try and hope it works, I will store your schematic in a safe place.


Author:  mwhouston [ 21 Dec 2013, 00:57 ]
Post subject:  Re: 6EM7 power amp

I see Matt has increased the inter-stage cap up from 0.022uf to 0.047. I knew I could go larger than 0.022uf but went mean-way between the two limits. I would still suggest a Russian Teflon inter-stage. I originally had orange drops which I think we all agree are very good but the Teflons (first time I have ever used them) I KNOW are better. I think I paid less for these extremely heavy, steel cased Teflons than orange drops. If you go with 0.033uf you will have more bass than you can handle. My system, Abstinence goes down to 40hz easy. Bigger is not always better in this case.

Owing to the fact I bought a NOS\NIB pair of RCAs and 10 single Samsung 6EM7s I'm sure there is another amp out there. I have managed to get ride of 6 tubes so far ($6 each) and in the end I may just keep one pair. Maybe with the next amp I can sell it with spare tubes to use some up.

Author:  mwhouston [ 03 Mar 2014, 06:29 ]
Post subject:  Re: 6EM7 power amp

A friend did a frequency plot of Abstinence. Nice and flat all the way out. Right down to around 13Hz. Distortion levels were high at around 3%+

Author:  Geek [ 03 Mar 2014, 06:54 ]
Post subject:  Re: 6EM7 power amp

Sweet! :thumbsup:

3% is quite nice, since most triode designers rate their amps at 5 or 10.

Gotta love that tube, mate! :D

Author:  mwhouston [ 03 Mar 2014, 14:28 ]
Post subject:  Re: 6EM7 power amp

Geek wrote:
Sweet! :thumbsup:

3% is quite nice, since most triode designers rate their amps at 5 or 10.

Gotta love that tube, mate! :D

I thought it too high but you make me feel better now. It definitely does sound like 3%.

Author:  Geek [ 03 Mar 2014, 17:32 ]
Post subject:  Re: 6EM7 power amp

That's because SS marketing has told us that unless an amp has 0.00000000001% THD, it's garbage ;)

A great way to learn the sound of distortions is dial up a certain THD on a SET, SEP, PP triode and PP pentode and play a familiar piece (female vocals pref. because of spectrum).

3% on a SET is pretty easy on the ears because it'll be mostly 2H.


Page 24 of 25 All times are UTC - 6 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group