will this circuit work?

[yunger]

am i way off base with this? i scrapped the ef86 preamp of a dc30 for a simple 12ax7 preamp. i'm sure somebody has tried something like this before. i wonder how changing the second channel would affect the first.

[ETR]

I'd probably go with a bigger grid stopper on the second gain stage for one thing.

[Chuck H]

Thats almost a dead on copy of the "hand drawn" dc30 schem on Schematic Heaven. I know nothing about the dc30, but if you ask me that 220k plate load for parallel 12ax7 stages is too high. I've gotten a good voxy vibe using a single 12ax7 stage with a 220k plate load and a 2.7k/22uf cathode. For parallel stages that would equate to a 110k plate load and a 1.35k/44uf cathode. So I might try a 100k plate load and a 1.5/22uf cathode for parallel stages just because they are standard values.

It's worth noting that the 220k plate load on the parallel 12ax7 stages equates to a 440K plate load if you were to run a single 12ax7 stage. I've only seen that done once in a hack build and it sounded like caca. Since it's hand drawn I would question the validity of the dc30 schem on Schematic Heaven.

Oh, and FWIW the stock "Vox" tonestack in that "hand drawn" schem is a real piece of work. I don't even think Vox uses it anymore on the RI amps because it doesn't behave as it is supposed to. Also, I had better luck getting "Voxy" tone from a 12ax7 using 500k treble and bass pots and a 150p treble cap in a standard BF type tonestack.

Chuck


Chuck

[yunger]

etr, what would you suggest for the grid stopper? and for my education and any other newcomer into tube amps, what will this do?
chuck, why are those 12ax7 triodes paralleled to begin with? is there any benifit as opposed to just leaving one stage empty?
thanks for the advice guys.

[ETR]

etr, what would you suggest for the grid stopper? and for my education and any other newcomer into tube amps, what will this do?
chuck, why are those 12ax7 triodes paralleled to begin with? is there any benifit as opposed to just leaving one stage empty?
thanks for the advice guys.

Anywhere from 100K to 1M, I'd just try different values to hear what sounds best.

[Chuck H]

why are those 12ax7 triodes paralleled to begin with? is there any benifit as opposed to just leaving one stage empty?
thanks for the advice guys.

The two stages are parallel probably because Matchless used parallel 12ax7 stages in some amps. It does have some advantages. Lower noise and lower output impedance are two of the benefits.

I never leave a tube stage just "hanging". For some reason it just bothers me. If I have a spare triode in a design I usually use it for an effects loop, a cathode follower, a swtchable overdrive stage, a parallel stage or, there's this cool PI I designed that uses three triodes instead of two that provides lower output impedance than the typical LTP. It helps prevent grid loading under overdrive.

In the schem on the parallel stage cathode circuit there is a .1 cap parallel with the regular 25uf cap. IME you won't hear a bit of difference by doing this so don't bother. Just use the regular 25uf cap across the 1.5k cathode resistor. And make that cathode resistor a 1 watt rating or you'll cook it. If you do decide to try the 220k plate load be willing to experiment with swapping it for a 100k plate load and let your ears decide which is best.

Chuck

[tubeswell]

there's this cool PI I designed that uses three triodes instead of two that provides lower output impedance than the typical LTP. It helps prevent grid loading under overdrive.

Sounds interesting - can you please post your schematic then?

[yunger]

In the schem on the parallel stage cathode circuit there is a .1 cap parallel with the regular 25uf cap. IME you won't hear a bit of difference by doing this so don't bother. Just use the regular 25uf cap across the 1.5k cathode resistor. And make that cathode resistor a 1 watt rating or you'll cook it.

i wondered about that. i've seen it on every dc30 schematic.

looks like i have a lot of options to tweak with. should be lots of fun.

[Chuck H]

The poly cap in parallel with a bigger electrolytic is an old trick to get better performance from a mediocre electrolytic cap. It was used on alot of top end hi fi stuff "back in the day". Matchless no doubt used it because it was considered an "uptown" trick to maximise performance. Thats all well and good. But the performance of elecrolytic caps back in the day wasn't as good as it is now. Even then, though you could measure a difference you may have been hard pressed to hear one. With good quality low impedance modern electrolytics you really shouldn't need to bother with it. Then again, it's just one .1uf film cap to make the circuit "schematically correct". So though I say "why?" it's also fair to say "why not?"

Chuck

[Chuck H]

Sounds interesting - can you please post your schematic then?

Ok. But here's my disclaimer...

When I came up with this idea I used it in an experimental amp I was developing. The preamp didn't sound right and I was in the process of tweaking it when I got a call from a famous guy that wanted a sample amp with specific features. His time frame was tiny so I had to work with what I had on hand. That meant scrapping anything experimental and going with a design I already had. Also due to time constraints I had to use the chassis I was already working on. So I stripped it down to sockets and transformers and rebuilt it without the fancy PI circuit. So...

This circuit has only been minimally tested and only by ear, never on a scope. I have not had the opportunity to rebuild it yet. The good news it that it seemed to be working really well. In fact it's the only part of the experimental circuit I wasn't going to change. However, My claim that it reduces grid loading is only based on theory and listening tests. Not actual scope readings.

I can tell you that shunt feedback arrangements do have lower output impedances than the typical PI gain stages. Even though the circuit is AC coupled it still seems to provide a considerable improvement over the typical LTP for reducing grid conduction.

As mentioned above this circuit uses shunt feedback to reduce output impedance. If you need more gain then you would change the 22k input resistors on the shunt gain stages to a lower value. This would raise gain, but also output impedance. So it's a trade and balance thing. I was building EL84 amps at the time so I didn't need that much swing. But this curcuit should be able to provide enough voltage to clip most any standard guitar amp tube.

Chuck

[chipaudette]

Did you really intend to have a 50pF treble cap in Channel 1? Did you mean 500pF?

Chip

[yunger]

Did you really intend to have a 50pF treble cap in Channel 1? Did you mean 500pF?

i just copied that from matchless schematics. as a matter of fact, in4 different schematics, i saw between 47pf and 56pf.

i'm surprised that i'm not getting more comments on my channel 2 layout. that was the part i stuck in there in replacement of the ef86 channel in the dc30.

[Chuck H]

I was building EL84 amps at the time so I didn't need that much swing. But this curcuit should be able to provide enough voltage to clip most any standard guitar amp tube.

As a follow up...Here's how I would do it for the big bottles (EL34 / 6L6)

[Chuck H]

i'm surprised that i'm not getting more comments on my channel 2 layout. that was the part i stuck in there in replacement of the ef86 channel in the dc30.

Hmmm...

Ok, Well the Vox uses a fancy little phase correction circuit so the channels will be in phase at the output. The way you have it you can't jumper the channels without a problem. Other than that I would say it looks like a decent circuit. I might change a couple of values but nothing earth shaking. You will probably tweak it up before your done anyhow.

Chuck

[chipaudette]

I'm pretty much only used to seeing Fender/Marshall style tone stacks. The tone stack on channel 2 is very different from that, so I'm hesistant to offer suggestions or criticisms.

But, if I were to offer criticisms, I fear that you might have too much gain on channel 2. If that's what you're going for...then great. But, if you'd like to have a usuable clean range, I fear that this setup isn't going to do it. The use of two full-cathode-bypassed 12AX7's is a lot of gain. Sure, Fender sets up their 12AX7's just like you do, but they have a massively lossy tone stack in the middle. To my unsure eye, your tone controls don't look like it'll have nearly the Fender-level of losses. Therefore, I think that you'll have a lot of pre-amp distortion.

If you want to reduce the gain/distortion in this channel, i would do one or more of the following: (1) remove one of the cathode bypass caps...probably the one on the second pre-amp tube stage, (2) insert a 250K (or higher?) resistor right after your 0.22 uF coupling cap and before the volume pot.

If you want pre-amp distortion, then keep all the gain that you have...but I fear that you'll have too much bass that it'll cause flabby sounding distortion. To reduce bass, try one or more of the following: (1) reduce your 0.22uf coupling cap to 0.022uF or even 0.0022 uF, (2) reduce the value of the cathode bypass cap on the first pre-amp tube from 25uF to somewhere around 1uF (go higher or lower to taste).

Finally, seperate from the gain issue, I'm confused about your tone controls. Again, I'm only used to Fender/marshall tone stacks, so the one you have here might be perfectly standard. But, what confuses me is the connection (via the 0.01 uF cap) between the wiper of the volume pot and the top of your tone control pot. Why wire it this way? How is it supposed to sound? It appears to result in some looping around the volume pot that i don't fully understand...it makes me uncomfortable. If it is just a high-rolloff tone control, wouldn't it work just fine by removing this connection?

chip