Harmolodics, welcome to the forum. 😀
Hopefully Planet10 will be able to weigh in here on that mod, but you might also investigate the 6AB4/EC92 as a potential substitute for the 12AT7, this is a single triode that is virtually identical to one half of a 12AT7. Data sheet here: http://www.mif.pg.gda.pl/homepages/frank/sheets/049/6/6AB4.pdf
They're not particularly popular for audio use and shouldn't be too expensive. (And IME the quality of a good NOS example of any recognizable brand is going to be better than any reasonably priced EF86 you'll find these days)
http://www.ebay.com/sch/i.html?_from=R40&_trksid=p5197.m570.l1313&_nkw=6ab4&_sacat=293
Hopefully Planet10 will be able to weigh in here on that mod, but you might also investigate the 6AB4/EC92 as a potential substitute for the 12AT7, this is a single triode that is virtually identical to one half of a 12AT7. Data sheet here: http://www.mif.pg.gda.pl/homepages/frank/sheets/049/6/6AB4.pdf
They're not particularly popular for audio use and shouldn't be too expensive. (And IME the quality of a good NOS example of any recognizable brand is going to be better than any reasonably priced EF86 you'll find these days)
http://www.ebay.com/sch/i.html?_from=R40&_trksid=p5197.m570.l1313&_nkw=6ab4&_sacat=293
The JJ ECC81 is quite well priced, look to be cheaper than NOS 6AB4, and is regarded by some on this forum, such as Sy, to be one of the best ECC81 regardless of price. So that you are not 'wasting' one half of the tube, you could wire up opposite sides of the tube on different mono blocks. When you are reaching the end of the life of the triode section, swap the two tubes over and you have just got a 'free' replacement.
Last edited:
IMHO an excellent suggestion..
Attachments
The EF86 version shown will be an extremely bad choice due to severe distortion. It can be a little less if one removes the EF86 cathode bypass cap but still the EF86 is to wimpy.
Anode to anode/ Schade feedback works best with a pentode driver.
If you want an cheap pentode, use 6AU6. But don´t triode-strap it. This will outperform the original 12AT7 by far.
Another way to do it is to cascode the 12AT7. This will outperform any triodedriver.
The third way is to drop the tube driver and go for a depletion MOSFET. Just check Michael Kosters ingenious thread about this.
Anode to anode/ Schade feedback works best with a pentode driver.
If you want an cheap pentode, use 6AU6. But don´t triode-strap it. This will outperform the original 12AT7 by far.
Another way to do it is to cascode the 12AT7. This will outperform any triodedriver.
The third way is to drop the tube driver and go for a depletion MOSFET. Just check Michael Kosters ingenious thread about this.
There was the long running thread where people tried a pentode (6AU6) and reported back that they liked the ECC81 better. I could only attribute this to
- more second harmonic distortion
- lower damping with less control over the speaker and hence bass bloat/warmth
Personally I agree with you that a pentode driver has to sound better - and my experience is that it does. However it must be a high current pentode capable of at least 6mA standing current.
Some people have different expectations of what a valve amp is supposed to sound like.
Shoog
Hey Shoog,
I am not sure if I understand you right but I guess you mean the AT7 gave the above "symptoms" with higher 2ndorder and bloat?
I know you have played a lot with 6AU6. To me 6AU6 outperforms the original by far both when measuring and listening. But off course the RH needs adjustment on more component values to do better.
But I agree 6AU6 is suboptimal. If going for a D3a or E280 we are close to the goal.
Personally i would go for a DN2540(or some IXYS depl.fet) as this sandy thing outperforms all of the above and also makes the circuit a lot simpler.
I am not sure if I understand you right but I guess you mean the AT7 gave the above "symptoms" with higher 2ndorder and bloat?
I know you have played a lot with 6AU6. To me 6AU6 outperforms the original by far both when measuring and listening. But off course the RH needs adjustment on more component values to do better.
But I agree 6AU6 is suboptimal. If going for a D3a or E280 we are close to the goal.
Personally i would go for a DN2540(or some IXYS depl.fet) as this sandy thing outperforms all of the above and also makes the circuit a lot simpler.
I built an all-Octal version of an RH84 amplifier, using a single 6SL7GT driver and a pair of 6V6GT output tubes.
Rectifier tube was a dirt-cheap 6BY5G (coke bottle shaped) dual diode TV damper.....glows REAL nice too!
Steve
Rectifier tube was a dirt-cheap 6BY5G (coke bottle shaped) dual diode TV damper.....glows REAL nice too!
Steve
I think I'll go with Chris's idea and keep it simple as this is my first tube project. Though cascading the 12at7 does sound intriguing. I'll have to read up on how that works.
So about the power supply, am I right in thinking that the secondaries on my transformer (6.3V and 5.7V, with 1.6A or less) are incompatible with the 5u4? Originally, the M-5s used a 6x4 rectifier. Can that be dropped into the published RH84 supply without a fuss? (someone seems to have done something similar here, post 8)
So about the power supply, am I right in thinking that the secondaries on my transformer (6.3V and 5.7V, with 1.6A or less) are incompatible with the 5u4? Originally, the M-5s used a 6x4 rectifier. Can that be dropped into the published RH84 supply without a fuss? (someone seems to have done something similar here, post 8)
Cascoding is after all not a good idea as it needs 400V B+.
Your PT will work just fine. You can take out more power than the RH with proper biasing as the RH is under-biased resulting in unsymmetric clipping. Use a GZ34/5AR4 if you want higher B+. You can also use your 6X4 without problems.
To refer to SY about 12AT7 being so great is in this case wrong as he uses them at totally different workingpoint at very low Ia where he found them very linear.
Even if triodes are suboptimal with Schade/Anode-Anode, a linear one to try could be the 6072A/12AY7.
Your PT will work just fine. You can take out more power than the RH with proper biasing as the RH is under-biased resulting in unsymmetric clipping. Use a GZ34/5AR4 if you want higher B+. You can also use your 6X4 without problems.
To refer to SY about 12AT7 being so great is in this case wrong as he uses them at totally different workingpoint at very low Ia where he found them very linear.
Even if triodes are suboptimal with Schade/Anode-Anode, a linear one to try could be the 6072A/12AY7.
Last edited:
Are you referring to these measurements?
http://www.diyaudio.com/forums/tubes-valves/164920-guess-tube.html
This was done with a 12AX7 - with this tube it makes completely sense why one would want to use a very light load for linearity. Or did SY do similar measurements with an 12AT7 that I missed?
I also remember that he recently stated that he prefers JJ 12AT7 over most (more expensive) NOS versions as mentioned by Crish. See here:
http://www.diyaudio.com/forums/tubes-valves/190075-12at7-12au7.html
Cheers,
Martin
Hey Martin,
Apparently you have missed it😉. Don´t think we´re talking 12AX7 here. But the same about light load actually goes for 12AT7.
About 12AT7, it was the info presented by SY in his Red Light District EL84 amp. Think I also discussed it with him in another thread a few years ago. Just make a loadline in a Ua/Ia with B+ 400V, Ua 110V and Ia 1,8-2mA to check out.
So the 12AT7 is used with just a slightly sloping loadline where in a RH it´s close to vertical. Load as seen from the RH driver is in the ballpark of 2-5kohm.
From the above we can see the driver should ideally be a AC-current source ie. pentode, cascode or FET.
Apparently you have missed it😉. Don´t think we´re talking 12AX7 here. But the same about light load actually goes for 12AT7.
About 12AT7, it was the info presented by SY in his Red Light District EL84 amp. Think I also discussed it with him in another thread a few years ago. Just make a loadline in a Ua/Ia with B+ 400V, Ua 110V and Ia 1,8-2mA to check out.
So the 12AT7 is used with just a slightly sloping loadline where in a RH it´s close to vertical. Load as seen from the RH driver is in the ballpark of 2-5kohm.
From the above we can see the driver should ideally be a AC-current source ie. pentode, cascode or FET.
Yes, it seems I really missed that post. 😀
The point of better linearity with a lighter load is well taken and I won't argue about that.
I was just impressed how better these figures can get in the case of the 12AX7 when I looked at the data that SY presented in his "Guess a tube" thread. I was not sure if the effect was as dramatic for the 12AT7 as well. Therefore, I was confused if you were probably referring to the 12AX7 post.
That's why I asked (it was a genuine question).
I also won't argue about the heavy load for the 12AT7 in the RH84 circuit. Indeed, I am not even sure that linearity was the foremost design goal in that circuit. A good dose of even harmonics might have been intended. After all, some of the builders report that they like the sound very much.
Just for the record: I have not built this amp myself. My main system uses a slightly modified version of Gingertube's Baby Huey (PP EL84 with Schade style feedback - I like that one very much but it's a completely different beast).
Cheers,
Martin
The point of better linearity with a lighter load is well taken and I won't argue about that.
I was just impressed how better these figures can get in the case of the 12AX7 when I looked at the data that SY presented in his "Guess a tube" thread. I was not sure if the effect was as dramatic for the 12AT7 as well. Therefore, I was confused if you were probably referring to the 12AX7 post.
That's why I asked (it was a genuine question).
I also won't argue about the heavy load for the 12AT7 in the RH84 circuit. Indeed, I am not even sure that linearity was the foremost design goal in that circuit. A good dose of even harmonics might have been intended. After all, some of the builders report that they like the sound very much.
Just for the record: I have not built this amp myself. My main system uses a slightly modified version of Gingertube's Baby Huey (PP EL84 with Schade style feedback - I like that one very much but it's a completely different beast).
Cheers,
Martin
Hello Martin,
You´re probably right about even order harmonics, as also Shoog said:
- more second harmonic distortion
- lower damping with less control over the speaker and hence bass bloat/warmth
Nice´n cosy!
You´re probably right about even order harmonics, as also Shoog said:
- more second harmonic distortion
- lower damping with less control over the speaker and hence bass bloat/warmth
Nice´n cosy!
I just rewired original Almarro A205A MKII to RH84 style amp. Now it sounds much better.
The most funny thing is that almost all necessary parts are reused from stock amp🙂
The most funny thing is that almost all necessary parts are reused from stock amp🙂
I don´t know how Hammond tests their transformers but when I measured an 807 SE amp with 125CSE OPTs I got something like 35Hz to 30kHz @ 1,5W.
Finally getting my mono RH84s soldered together. Took forever to build the chassis. But they should look nice in walnut and aluminum.
One quick questions for whomever's around: The salvaged Akai M5 power transformers I'm using have two primary inputs, 110v and 117v. Being in the US, I assume the 117v is the correct one to use. Is it a viable idea to wire the mains to the 110v instead? Not sure I fully understand this, but would this increase the secondary voltages slightly? Is it a good idea?(since the M5 secondaries are 250v and the RH84 calls for 300v)
One quick questions for whomever's around: The salvaged Akai M5 power transformers I'm using have two primary inputs, 110v and 117v. Being in the US, I assume the 117v is the correct one to use. Is it a viable idea to wire the mains to the 110v instead? Not sure I fully understand this, but would this increase the secondary voltages slightly? Is it a good idea?(since the M5 secondaries are 250v and the RH84 calls for 300v)
Seems it would be great for the plate voltage, but your filaments would be higher too... probably not that high, I'd go for it.
Hi Harmolodics,
I don't want to curb your experimental curiosity but do you think the output voltages of your power transformers need elevating?
In the RH84 schematic, a B+ of around 300Vdc is called for. Mr Kitic has chosen a 300-0-300 transformer to supply this but it's not the only way to go. 250 Vac rectified gives around 350Vdc. This will be pulled down by valve rectification, ac ripple filtering, load etc. Using solid state diodes instead of the 6x4 valve would yield a higher B+ if needed. Let us know if your committed to the 6x4 valve route and maybe we can tailor the schematic to suit.
Ps what else are you using from the Akai. Chokes?
Brgds Bill
I don't want to curb your experimental curiosity but do you think the output voltages of your power transformers need elevating?
In the RH84 schematic, a B+ of around 300Vdc is called for. Mr Kitic has chosen a 300-0-300 transformer to supply this but it's not the only way to go. 250 Vac rectified gives around 350Vdc. This will be pulled down by valve rectification, ac ripple filtering, load etc. Using solid state diodes instead of the 6x4 valve would yield a higher B+ if needed. Let us know if your committed to the 6x4 valve route and maybe we can tailor the schematic to suit.
Ps what else are you using from the Akai. Chokes?
Brgds Bill
Last edited:
I'm using all of the Akai iron (power, choke and output). I've found a few other instances on this forum of people using the same salvaged transformers for RH84 builds and the consensus seems to be that it works well. What would you suggest altering from the original schematic?
- Status
- Not open for further replies.