Express Combo using 6V6s (Finished)

[UR12]

I have been working on an Express combo for a few months now...actually my trusty amp tech is doing the electronics build with an Allyn chassis while I am constructing the 1x12 cab out of flame maple (all edges are box jointed, dimensions match my Bad Cat Hot Cat 30). The face plate is a 3mm piece of polished copper that I had the local trophy shop engrave the lettering in. I then filled the engraving with .925 silver. I found 3 NOS early 80's Celestions in a friends basement so that is the current speaker of choice. The amp is nearing completion soon...I will post photos as soon as the cab is more complete.

Sounds cool . Can't wait to see some pics of the finished amp. And that friend's basement.............we should all have friends like that

[M Fowler]

Dana,

Good thing I reread this thread because you and Billy answered what I wanted to know about cleaning the Express up some so I can build myself a versitile combo amp with 212 celestions. Dana I got your chassis a while back and now I realize that I should just build the Express rather than a Bassman or JTM 45. But I was worried about having some clean room before the well known express gain kicks in. Dana noted to use a 68K NFB and Billy 100k. Any other advice guys before I get started? Oh also on that VVR is that useful considering the breakup of the Express?

Thanks.

Mark

[UR12]

Dana,

Good thing I reread this thread because you and Billy answered what I wanted to know about cleaning the Express up some so I can build myself a versitile combo amp with 212 celestions. Dana I got your chassis a while back and now I realize that I should just build the Express rather than a Bassman or JTM 45. But I was worried about having some clean room before the well known express gain kicks in. Dana noted to use a 68K NFB and Billy 100k. Any other advice guys before I get started? Oh also on that VVR is that useful considering the breakup of the Express?

Thanks.

Mark

Mark there are a few things you can do to clean the amp up. You can also adjust the size of the 150k resistor and coupling cap (.002) on the grid of the 3rd gain stage. I have actually used a trim pot in place of the 150k and adjusted it for the sweet spot. I put a VVR on ever amp I build unless it has a reverb or tremelo.

[v00d00blues79]

Dana,

Those amps look freakin' sweet! Question: are those cabs the same size as, lets say, a Marshall BB combo cab? I have one here that's not doing anything and this would be awesome to drop in there.

Thanks,

Andy

[UR12]

Dana,

Those amps look freakin' sweet! Question: are those cabs the same size as, lets say, a Marshall BB combo cab? I have one here that's not doing anything and this would be awesome to drop in there.

Thanks,

Andy

Yea it is the same cab I use for the 2 x 12 Marshall 18 watt. I't cool cause you end up with a wolf in sheeps clothing

[M Fowler]

I would like to go with the EL34 rather than 6V6 do you think that will be a problem with the tight confinment of the combo?

I have a bunch of 6L6s sitting around so I had thought of doing a Bassman type of amp but the more I think about your build it gets me fired up to try it.

Thanks Dana

Mark

[billyz]

Regarding what I said about the NFB resistor. 100k will not get you a cleaner sound. You would go to a lower value like Dana mentioned. The 150k reduced to a lower value will throw away some more gain too, equals more clean.

El34's need very little more room than 6v6's, but I would opt for a larger cabinet and 2x12 speakers. Your gonna push alot more air and vibration, so it could be more problematic than a 6v6 combo in that regard.

You could get more headroom with a 5751 in V1 and or V2, I find the 12at7 in V3 is fatter and not as crunchy as the 12ax7, less bright too.

And definitely use the Spit plate load circuit to get more clean headroom.

Dana makes a remote switching circuit for it too, I believe.

[rooster]

OK, lets clear this up a bit?

The 100K is the stock Express value and billyz runs the stock value. Since the NFB resistor, circuit-wise, is looking at the 5K Presence pot as the ground of the PI tail, its the ratio of these two components that determine how much energy from the OT is fed back to the PI. (BTW, this energy is now retuning to the circuit with a reversed polarity of what it was when it originated - important!) Which in turn either helps tighten things up (more signal fed back>smaller ratio), or loosens things up (less signal fed back>larger ratio).

(Also, when billyz says he prefers amps 'without any feedback', he is saying he likes an amp that doesn't have this circuit added. An AC30 is a good example of such an amp, as are a lot of older Class A amps, P/P or SE.)

So 100K/5K = 20:1 ratio, and 68K/5K = 13.6:1 ratio.

This means that increasing the NFB resistor to 200K would make the amp more 'ragged/loose' than the stock 100K value. Which means that the same amp with a 68K NFB resistor vs. a 100K value would be 'tighter' overall.

So when billyz says: "Regarding what I said about the NFB resistor. 100k will not get you a cleaner sound. You would go to a lower value like Dana mentioned." - what he is saying is that if you want a 'tighter' sounding amp, make the ratio of these two components smaller....

The component that billyz does talk about changing is the 'tail' of PI. He changes the stock value, 10K, to 22K. This change will make the PI have less 'gain' than stock. If you went to a 47K value, you would have less gain still, given the same PI.

OK, since we are here... Going back to the AC30, it has no negative feedback circuit. So here's an option for a kind of 'loose' sounding amp. But then its PI has the 47K resistor tail creating less gain than 10K resistor would in the same circuit. ?? Yeah, eventually came the TB channel, bumping up the preamp and making it something many people think is very cool. But before this, in its middle period, the non-TB version AC30/6 was a really quiet and also somewhat loose amp. Interesting combination and, of course, definitely the kind of girl Brian May wanted to take to the Prom...........

OK, you are bored, but I feel better.

[billyz]

Rooster, Thanks for the detailed explanation and you are exactly correct. I was being too informal.

Since this is the TW Discussion, I assumed any deviation would be from the TW circuit. Though I have seen that NFB loop in some old Marshall's and old Fender's . And for the record, my favorite amp without NFB is the late 50's Tweed Deluxe, Class a/b, cathode biased, cathodyne phase splitter.

Anyway, Thanks again. This is a great place with a lot of really knowledgeable and gracious people. How come there aren't more women ampoholics ? never mind.

[M Fowler]

Thanks to all of you for giving me a very good lesson I appreciate the efforts taken here to help out, very informative.

Mark

[UR12]

One more thing worth noting is when you are changing the circuits and going from a power amp that is putting out 36 watts to one that is putting out 22 watts (EL34 to 6V6 for example), you will also have less negative feedback signal generated at the speaker side of the amp and thus less NFB. To compensate you can cahnge the value of the NFB resistor to something lower in value to let more NFB signal come through. For you guys building a 100 watt Express you may need to increase tha value of the stock NFB resistor. It is definately something you can play with and you may find that it not only makes the amp sound a little different it may also "Feel"a little different.

[rooster]

Dana presents an interesting consideration.

However, looking at the Fender Black Face series, given the identical AB763 circuit as an example, all the 6L6 equipped amps, including the BFTwinR run the 820/100 ohm combination or 8.2:1 ratio, while the 6V6 equipped BFDR changes this to an 820/47 ohm combination, or a 17.4:1 ratio. ......Which would say that, if you took Fender's advice here, you would increase the ratio, not decrease it.

So Dana you brought up a point, but I wonder if you have tried a higher value feedback resistor? I run the stock value, 100K, but now that you have made me take a look at this again, I think I will try a higher value. Following Fender's lead, 200K would be in order, but I think I will start at 150K to see if I can sense something different.

I'll kick this around tomorrow and get back here. billyz, pay attention...

[billyz]

I have used much higher ratios on the NFB circuit even a variable " soul" control that went from 15K to 150K. For myself I found this useful, but for the general public, it just confused them. So now I just find a happy medium and leave it fixed. I sell more amps .

But, I think everyone who builds should find their own circuits or variations of a proven design. I used have variable pots and switches all over the circuit, so that I could discover the full range of effect each part played.

The Ratios Fender used were all over the map,Looking at his tweed and brown/blond series, I am not sure what his design criteria was except sell more of what they want.

I will as you say " pay attention" Because people on this forum and others do not get glazed over looks on their face when discussing Amplifiers and there inner workings.

Peace or piece...........which ever you prefer.

[UR12]

Dana presents an interesting consideration.

However, looking at the Fender Black Face series, given the identical AB763 circuit as an example, all the 6L6 equipped amps, including the BFTwinR run the 820/100 ohm combination or 8.2:1 ratio, while the 6V6 equipped BFDR changes this to an 820/47 ohm combination, or a 17.4:1 ratio. ......Which would say that, if you took Fender's advice here, you would increase the ratio, not decrease it.

So Dana you brought up a point, but I wonder if you have tried a higher value feedback resistor? I run the stock value, 100K, but now that you have made me take a look at this again, I think I will try a higher value. Following Fender's lead, 200K would be in order, but I think I will start at 150K to see if I can sense something different.


I'll kick this around tomorrow and get back here. billyz, pay attention...

hey Rooster

Yea I have tried the 100k in this amp along with a few other values. Fender uses the 100 ohm on it's 100 watt amps and the 47 on the 50 watt amps like the Showman and also on the BFDR using 6v6s. By changing the PI tail value you are going the opposite direction compared to changing the NFB resistor value. Going smaller with the PI tail resistor value would present less NFB and increase gain which is the same as going with a larger value NFB resistor. Since there is no presence control you have the option of going either way to change the ratio you speak of. You can also go with the alternate presence control using a 25k pot that is not in the DC path of the PI cathode. This method substitutes a 4.7k - 5.1 k resistor for the presence pot and then you could play with that value while keeping the presence control working.

[rooster]

Dana - OK, trying to keep a consistant language here, the PI 'tail' in the Express and Fender BF amps refers to the resistance to ground that comes off the juncture of the PI grid and cathode resistors, i.e., the two 1 megs and the 470 ohm resistor.

So in part, this would be the 10K resistor on the Express, which is where the OT feedback is returned, and then the added resistance to ground - off of that - which is the 5K Presence pot. BTW, I think it is because there are two resistors here, in series, with the feedback voltage returning at a point between the two, that it is named the "long tailed pair". So the two resistors together are the actual 'tail'. And actually I think that if it is to be considered a true 'long tailed pair', there needs to be the OT feedback applied.

Translating then, the VoxAC30 is refered to at times as having a version of the 'long tailed PI' - but I don't think it can really be considered a true example because the negative feedback is not applied..

Anyway, to finish up, as far as I understand things, the 10K value can be increased with mainly the affect of balancing the PI and also reducing gain - and the higher value resisitor, say 47K would also reduce the negative feedback voltage from reaching the PI more than a 10K value, true enough. .........However, this being said, the ratio of NFB voltage is still determined by where it sits in relation to ground, and this is determind by the second resistor in the 'long tailed pair', which in this case is the 5K pot, (or the 47/100 ohm resistor in the BF Fender series.)

So when you determine the *ratio* of NFB in the PI circuit, you still need to divide the resistance coming off the OT secondary by the second resistor in the long tailed. Then given the constant first resisitor in the long tailed pair, you can start talking 'NFB ratios'.

I think a cat can be skinned many ways, true, and you can change that first resistor in the tail and it will affect things that involve the NFB. ...But for the sake of speaking and understanding things, there still is a ratio and it has nothing to do with the first resistor in the tail. Are we on the same page here?