I didn't know that information, thanks. The 102d though is spec'd 2.1v at 1a on the filaments. The 26 seems also a good valve for the application (I've just had a look at the data sheet) and the only reason really is that I already have the 102d.
The similarity I was drawing upon was more of what I observed with the 2a3. That is, compared to my vintage RCA 2a3, both my Shuguang and Full Music modern production ones are quieter. (Also, the globe shape which has been said is noisier is actually not the case between the two Chinese examples.)I'm hoping that this would be the case with a modern reproduction 102d.
My feeling is that a lot of current manufactarers are republishing original tech specs rather than ascertaining what the real characteristics of their modern reproduction are like. For example, it has been reported of Sovtek 300b running away at 400v Va when the max is claimed at 450v. I myself have operated the 2a3c with about 330v, -60v, 70ma so 23w dissipation. Low 300b territory, nothing untowards with the valves I have anyway. (I don't advocate of course, it was tried with just one set as example.). I'm rambling because I'm trying to be optimistic
.
Ah - so it is, I mis-read it at 340mA, which I can only imagine was because the plate current is 34mA at the operating point. Should be fine then. Looks like a nice tube.
I'm doing well, aren't I...... I have a presentation tomorrow on my mind! Trying to work for it rather fitfully and distracting myself with other things.
For a split second there I thought I'd mixed up the spec sheets with a 205D or something I went cold for a bit . We'll see how it goes with this valve anyway.
Now that it seems feasible to build the output stage separately, as I was imagining, then I'll probably be trying higher voltages on the 845 sooner rather than later. It was the idea of adjusting the values of the input/driver stage components every time sizeable increases in B+ was made was what was making me weary.
I'm more excited now
I went cold for a bit . We'll see how it goes with this valve anyway.Now that it seems feasible to build the output stage separately, as I was imagining, then I'll probably be trying higher voltages on the 845 sooner rather than later. It was the idea of adjusting the values of the input/driver stage components every time sizeable increases in B+ was made was what was making me weary.
I'm more excited now
Hi,
For some time now, I too have been thinking of building an amplifier with two separate stages, each with its own power supply. The first stage with the driver tube ends with the secondary of the interstage transformer. The second stage with the power tube starts with the grid connection of the 845 (or other tube) and ends with the speaker outputs.
However, I have a number of questions.
1 - can the first stage (but also the second stage) be damaged if the connection between the two fails when in operation?
- the second stage is connected to the first stage probably with a 2-pin cable (ground and signal): would a normal cable with rca pins (like the ones we normally use to connect the cd to an amplifier) be OK? and if so, of what length? (friends advise me against this design precisely because of this, as the connection between the interstage and the power amp grid had better be as short as possible)
- for the second stage the signal I think necessarily needs a reference to ground consisting of a resistor (of what value??) or a grid inductance, when - if everything were in one chassis - there would be no need for this as the interstage transformer (if carefully chosen) does not need any load resistance on its secondary before entering the grid of the power amp.
Thanks
That's a very old post! I never did build an 845 amp. I use a 2a3 output stage. I do have a driver stage in another chassis, though, like you say. I don't use an interstage, just cap coupled so it's just an ordinary RCA lead. You can put 1M or 2M on the output of the driver stage and then 100K on the input of the power stage, the 2a3 in my case. So just 2 DHT stages. It needs a step-up transformer in front of the driver stage, which is a 10Y. I use a Hammond 1140LN-C in 1:4. Maybe someone else can advise on using an interstage.
I sometimes use a separate B+ supply for the driver stage but it's a bit wasteful. Easier to get the B+ off the power stage.
I sometimes use a separate B+ supply for the driver stage but it's a bit wasteful. Easier to get the B+ off the power stage.
Using an RCA connector for voltage is just asking for deadly trouble! An uninformed, well meaning idiot could plug anything that has an RCA on it into high voltage! This might end in death of both the amp and the idiot! Always use dome sort of locking connector that is not used anywhere else to avoid mistakes. Companies like Cimch make great mil spec connectors with as many pins as you need. Get the ones with screw rings to avoid accidental disconnections. Orient them so the female has the voltage on it so you cant accidentally touch a hot pin. Always plan for the worst possible event to happen and design to avoid it! It’s hard to find very high voltage wire in the audio world so make sure you have wire that is rated for it. Neon sign wire can work. Don’t cut corners because nothing is worse than someone dying because you saved a dollar unless you really don’t like them.
In the connection between the interstage transformer (which is in the first chassis) and the power valve grid (which is in the second chassis) no voltage passes, only the signal. Thus - of the two cables (see diagram #29) connecting the two chassis, one is the cable that joins the grounds of the 2 chassis, while the other only lets the audio signal pass between the secondary of the interstage and the grid of the power tube. The power supplies of the two chassis are each independent in their own chassis and there is no passage of dangerous voltage between one and the other.
I fail to see where you see the danger.
Special connectors capable of carrying high voltages between separate chassis are normally used when - for reasons of space - the amplifier is built with the power supply on one side and the valves and interstage and output transformers on the other.
But this is not the case here.
Of course you don't use an RCA connector for voltage but that isn't what I said or would ever say. The RCA connector just connects the signal, as it's always done. You would use a power connector for voltage.
It seems logical to me that an approach of the kind shown here has as its only advantage the modularity of the amplifier.
Having the driver and power tube in separate chassis makes it possible to try different drivers on the same preferred power tube. Or use the same driver on several other power tubes to listen for sonic differences.
Having 2 separate power supplies is uneconomical and would make little sense: unless it also makes the amplifier perform better electrically and sonically.
At least that's how I see the prospect of this possibility.
Having the driver and power tube in separate chassis makes it possible to try different drivers on the same preferred power tube. Or use the same driver on several other power tubes to listen for sonic differences.
Having 2 separate power supplies is uneconomical and would make little sense: unless it also makes the amplifier perform better electrically and sonically.
At least that's how I see the prospect of this possibility.
Yes - that's how I see it. Smaller chassis linked up. One power supply, one output stage, one driver stage with SUT if needed like 1:4. There is your amp. I did it that way to try several different driver tubes plus some outputs. The other reason was ease of handling. Each chassis was lighter and easy to move around and work on, given that they were constantly going from the bench into the audio system. The filament supplies are in each chassis with their DHT tubes, in filament bias where possible. You could put the outputs and the PSU together in a larger chassis. That works too. Filament supplies for DHTs are quite large, plus heatsinks and Rod Coleman regs, so you would run out of space on one chassis.