• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Tube designs: 45 driving 45/2A3/300B?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi everyone,

I've been thinking of going all out and building a super expensive low power (SET perhaps?) tube amp with 45 tubes in driver position, powering 45, 2a3, or 300b. I've seen some designs with the 45 as output tubes, but I haven't found one with it in driver position. Does anyone have any ideas?

Cheers and thanks!
 
Does anyone have any ideas?
45s are scarce and expensive. Also, there is the issue of controlling residual hum levels. Check out the curves for triode wired 6V6s. They give away NOTHING in the linearity dept. to the revered 6SN7. Use a triode strapped 6V6 to drive the DHT "final".

BTW, the 6AQ5 is essentially a 6V6 in a 7 pin mini bottle. It would kick butt as your driver.
 
do you understand gain and grid current?

Do you know how to supply and limit grid current in case of over power of the 300B???

Do you know about gain requirement, feedback used, to be able to have a proper amplification of the signal, not too much, not too little.?

if it is no to anyone of the 3 questions, you better start off with a kit, a proven design, instead of trying your own.

You should look at tubelab boards, this will get you a nice start, a solid plan proven design, and performance.
 
Hi everyone,

I've been thinking of going all out and building a super expensive low power (SET perhaps?) tube amp with 45 tubes in driver position, powering 45, 2a3, or 300b. I've seen some designs with the 45 as output tubes, but I haven't found one with it in driver position. Does anyone have any ideas?

Cheers and thanks!

I'm with Eli Duttman. Years ago I tried every sort of combination. You really want a clear, firm, punchy-but-sweet driver for a 300B or 845, and the 45 isn't it. Do a google search for "voyager 300B schematic". It uses a triode-wired EF86 amplification stage (gain of 35) and a triode-wired 6V6 driver. For sheer beauty you can't beat it. Just my two cents.
 
I built with 71A driver to 2A3/300B (JJ type 2A3-40W mostly 300Vp-k 65mA into 3k3), SE IT to PP output and also direct coupled in SE. I wouldn't go with 45 due to filament requirements, 71A is 250mADC, and you can filament bias the stage with 150R unbypassed cathode resistor for ~2k5 drive which is fine, or you can ground the filament return and DC couple the input tube.

Vp 180V, Ip 20mA, Vg -40Vg, Mu 3.

IME the input tube determines the overall sound, the driver and output are linear.. if you use a linear input tube, flavour will then be with the phono stage. Add colour to taste, and IME that is a good way to do things. Simple changes up front are easier to hear. Saves you going crazy :).. where's that salt..

HK
 
Last edited:
He did not replied, maybe a newbie,

I was going to suggest the 'new' ecc99 , it is made for that purpose.

It is the most economic linear triode ever to drive quite substantial loads in anode...

I did a comparison with an ecc99 with a EL822 at around 15 ma for the whole LTP, versus the 6sn7 with a transistor regulator Z something at 8.5 ma, the ECC99 was sounding as good than the best 6sn7 NOS I could throw at it.

The advantage of the ecc99 is the low plate resistance similar to a 45 tube or something like that.

The only disadvantage, I read this tube starts to wear very fast at around 1000 hour maximum useful life. Around 24$ for matched section, 800ma filament, this is a lot better than a big 45 and the expenses. It also uses a 9 pin socket, it is small.
 
Well it's not economically convenient if the useful life is just 1000 hours. You can buy a pair NOS/NIB 45's from reputable sellers at $ 160-200 and used as driver (something like 220-250V @25-30mA) it will easily last 10000 hours. Long life is one the many advantages of this old triode. The 45 needs more current but the consumed filament power is 7.5 VA (for a pair) vs 5 VA of one ECC99. Not a big deal really.
 
I know very well as I have done it but they don't seem to have any....I told them elsewhere as well

One thing I have to say is that 71a needs more care about microphony. The 45 is better in this department. Actually is probably the best among small power triodes.
 
I know very well as I have done it but they don't seem to have any....

Great tube: lower input capacities, 250mA filament current, 2k Rp, can swing linear 200Vp-p with ~20% headroom when biased at -40Vg. IT loaded ~190V supply at 20mA. Easy to filament bias and I think better sound than 45. $100pr eBay, they made very many of these tubes. Lots around.

Re microphony: construction of globe types with mica supports over the glass arbour types does help. All ST types will be with mica supports. In any event, I've not had a problem with any of them. And for an extra 250mA filament current you can try the thoriated tungsten filament type.. lots to like.

I'm not sure much comes close to 71A when operated within its limits.
HK
 
Last edited:
I think better sound than 45.
That's personal I suppose.

$100pr eBay, they made very many of these tubes. Lots around.
I know although sometimes one has to buy individual tubes to make pairs of identical tubes. This true for most old triodes now. Those sold as matched pair of identical tubes are getting more expensive.
Anyway I was referring to the OP that already has got the 45's and should buy the 71a instead.

Re microphony: construction of globe types with mica supports over the glass arbour types does help. All ST types will be with mica supports.
That's the beauty of the 45. You don't need to buy the more expensive globe type. It is possibly the quietest of power tubes. No surprise about the fact that it "sounds great"...
As the driver is supposed to work away from grid current in all conditions the bias method is not a problem. Cathode bias works just fine. It is more important the filament supply. DC supply at the very least. Better if voltage regulated and even better if voltage and current regulated.

For me filament bias makes more sense with battery tubes where power waste is really small. Personal opinion, of course.
 
Apologies 45: I cant find the time to work out how to do the multi-quote thing. For the most part we seem to agree, just some minor displacement and with nothing untoward.

1. Of course the subjective bias, worthy to note is that my moniker is not 71A : ).
2. ST versions of 71A are all with mica support and are less microphonic than their globe alternatives (if that was ever a problem that could not be mitigated).
3. 71A are less expensive than 45's, and matching is 'just one of them things'. Matched, how, 'at where' and for what.
4. 71A was designed for battery: 71A - Everything2.com and as such is prime candidate for filament bias.
5. 71A was originally made with a thoriated tungsten filament and this can provide for some fun if/when it gets boring.

Kind regards,
HK

I believe that 71A is a better driver for 300B, stack them 3 high in a DC couple or IT the driver 1:1 bifilar to the output. The input capacitance, power consumption, filament arrangement: particularly WRT cathode/filament bias are all in favour. If IT coupling, with the lower Vp-k of 71A and with a stacked (DC coupled) input/driver, bifilar wound IT coupling is an option, whereas the higher voltage needed to energise the plate of Type 45 works against the voltage restrictions for pri:sec on that type of IT.

Cost of the tubes themselves in any event and within context of the build should be considered as negligible. Replacements less so, but there are many.
 
3. 71A are less expensive than 45's, and matching is 'just one of them things'. Matched, how, 'at where' and for what.
As I said, no question about that but that's not the point. The point is that the OP has to buy the 71a's while already having the 45's in hand. So in this case the 45 is "cheaper".

The 45 has another advantage. At later stage if one wants can buy new production tubes. The KR-45 is my favourite.

I believe that 71A is a better driver for 300B, stack them 3 high in a DC couple or IT the driver 1:1 bifilar to the output. The input capacitance, power consumption, filament arrangement: particularly WRT cathode/filament bias are all in favour. If IT coupling, with the lower Vp-k of 71A and with a stacked (DC coupled) input/driver, bifilar wound IT coupling is an option, whereas the higher voltage needed to energise the plate of Type 45 works against the voltage restrictions for pri:sec on that type of IT.

Cost of the tubes themselves in any event and within context of the build should be considered as negligible. Replacements less so, but there are many.
I have used both. The 71a is more cost effective but I prefer the 45. I really do not care about few VA difference in filament power. I simply don't waste it in filament bias and that makes things even. Bifilar OT is not an option for me. It's the best.

DC coupling to a 300 B can be an alternative for me but not stacking power supplies. It's by means of cathode (or source) follower with its independent power supply. Old AN style in few words. That works very well and is safe in ANY event.....
 
As I said, no question about that but that's not the point. The point is that the OP has to buy the 71a's while already having the 45's in hand. So in this case the 45 is "cheaper".

The 45 has another advantage. At later stage if one wants can buy new production tubes. The KR-45 is my favourite.

I have used both. The 71a is more cost effective but I prefer the 45. I really do not care about few VA difference in filament power. I simply don't waste it in filament bias and that makes things even. Bifilar OT is not an option for me. It's the best.

DC coupling to a 300 B can be an alternative for me but not stacking power supplies. It's by means of cathode (or source) follower with its independent power supply. Old AN style in few words. That works very well and is safe in ANY event.....

Dear 45,

Its easier for me to type than do the multi split post thing. Your preference for 45 is noted. The 'designers' preference for 45 is also noted.

Convention specifies DC in Watts, and for DHT driver we need DC power; 45 filament = 3.75W, 71A = 1.25W. 300% increase is not trivial if we want to filament bias the tube. Consider:
Type 45: Vg-40V, Rk 21R, 0.035A Ip+1.5AIf = 71 W dissipation.
Type 71A:Vg-40V, Rk 150R, 0.020A Ip+0.25AIf = 10 W dissipation.

700% difference forces you into the need for regular Rk value and cathode bypass capacitor. Unless you want to get creative, and I do.. but the compromise does not justify the means :).

Whether you 'like' bifilar winding does not matter. A direct coupled input stage ~110V to 71A will allow for this as an option. Type 45 in the same situation will not allow for this.

AN cathode follower DC drive with split supplies and output stage filament return at 0V, yes. This is not what I am meaning at all.

What I am talking about with stacked supplies and DC coupling is totally different.

One could flip this around with an input transformer and have driver plate at -75V for direct couple to the 300B grid and with 0V filament return, the series resistance in the plate circuit will decouple the capacitance of the plate choke on the driver and also add to the load impedance, yet the advantages over the 45 even without resorting to this should now be clearer.

Kind regards,
HK
 
Last edited:
Dear 45,

Its easier for me to type than do the multi split post thing. Your preference for 45 is noted. The 'designers' preference for 45 is also noted.

Convention specifies DC in Watts, and for DHT driver we need DC power; 45 filament = 3.75W, 71A = 1.25W. 300% increase is not trivial if we want to filament bias the tube. Consider:
Type 45: Vg-40V, Rk 21R, 0.035A Ip+1.5AIf = 71 W dissipation.
Type 71A:Vg-40V, Rk 150R, 0.020A Ip+0.25AIf = 10 W dissipation.

700% difference forces you into the need for regular Rk value and cathode bypass capacitor. Unless you want to get creative, and I do.. but the compromise does not justify the means :).

Whether you 'like' bifilar winding does not matter. A direct coupled input stage ~110V to 71A will allow for this as an option. Type 45 in the same situation will not allow for this.

AN cathode follower DC drive with split supplies and output stage filament return at 0V, yes. This is not what I am meaning at all.

What I am talking about with stacked supplies and DC coupling is totally different.

One could flip this around with an input transformer and have driver plate at -75V for direct couple to the 300B grid and with 0V filament return, the series resistance in the plate circuit will decouple the capacitance of the plate choke on the driver and also add to the load impedance, yet the advantages over the 45 even without resorting to this should now be clearer.

Kind regards,
HK

I know, you don't need to repeat. It's just that I prefer other options.
 
I have tried filament bias and I haven't found the benefits people talk about.
That's why I consider it a solution for battery tubes only.
Cathode bias for a driver works fine. I have to say that often people chose the wrong value for the bypass capacitor. In case, if one doesn't want cathode bias, can use fixed bias for the driver too. Not a big deal and it doesn't necessarily need a dedicated secondary on the power transformer.

The dual supply for a DC coupled cathode follower doesn't need a center tap as well.

There are several ways to do things it depends on what one has in hands etc....

Input transformers: the best ones (with wide bandwidth) don't like significant DC voltage differential between primary and secondary. For some few volts is the max you can have. Check out....before burning them!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.