• 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.

cap coupling to interstage transformer or direct coupling?

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

I have built a pair mono block 45 SE amp with cap coupling. The circuit is base on 2A3 Ranking Gordon SE amp. It sounds good and I just wonder there is way to improve the sound by mod it from cap coupling to IT or direct coupling? Thanking in advance for your input!

Below is the schematic of the amp that I build
 

Attachments

  • IMG_0517.jpg
    IMG_0517.jpg
    613 KB · Views: 915
That's the Bugle circuit. Rankin also designed a direct coupled 45 circuit using 6BM8 as driver and current source published in Sound Practices that you can try. But the B+ needs to be raised to 450Vdc though. If you're using choke loaded power supply per Bugle schematic, changing it to cap loaded will increase the B+. The 6BM8 pentode section really is acting as a slow warm up before the triode section starts to draw current so the 45 grid is not slammed with high DC voltage. If you want to try octal tubes, the 6EM7 will give you similar result with minor tweaks.

I have built similar direct coupled 45 circuits before and it can sound really good, provided the output transformers are quality items.
 

Attachments

  • 23426942-81DD-44BC-87B0-2EB99F44735B.jpeg
    23426942-81DD-44BC-87B0-2EB99F44735B.jpeg
    352.9 KB · Views: 382
Last edited:
I am currently not using any SE amps. I have built 45 circuits using Tango XE-20S and many PP irons on the cheap such as irons from Sansui 1000A. The Tango is transparent on the top but bass shy. Heard Nature Sound transformers with permalloy core that sounded very natural. I used 6DL4 as driver direct coupled to 45 and sounded nice. Also 6SL7 is more lush. All of these projects were done many years ago. I am sure other members are more qualified to comment here.


PS, In reply, it's best not to quote the entire post to avoid redundant images. A little editing will work. Sorry, I am OCD like that.
 
Has any one out there built the gordon rankin constant current 45 amp schematic posted here? I am gonna breadboard it and see how it goes, I have the iron and tubes and caps. Wondering how the 20 turn pot plays into the design. Is it used to tune into the 180v at the plate? The meter's range is noted to be 100 VDC. I don't get it.
 
The easiest way to simplify the circuit in post #1 is to remove the top triode section of the 12AT7 with an IXYS 900V or 450V current source. And a stereo amp will need only one 12AT7, and you do not have to worry about the filament to cathode voltage rating of the (now removed) top 12AT7. I expect the sound quality to be very close to each other between the 12AT7 top triode circuit, and IXYS circuit.

As to the Pentode constant current circuit: The 180V is approximate. Change the 45, and it will be different. Change the B+, and it will be different. As was stated in post #6, adjust the potentiometer for 35V across the 1k resistor, the 45 will have 35mA.

Actually, one of the 45 data sheets suggested operating points is 36mA, and 275V plate to filament. It is the highest power out in the data sheet, but I do not remember the required primary impedance of the output transformer, I think it is 4600 Ohms. That should give 2 Watts (if the output transformer is low loss).
 
Last edited:
With IT coupling the frequency response and transient response get worse and hum/interference pick-up will increase. But I don't know about the sound improvement.

with IT coupling the frequency response and transient response improve and hum/interference pick-up decrease the sound improvement its great DHT triodes was made to work with transformers
 
the question its about interstage isolating coupling

try a bifiliar IT, no more of 250V between primary and secondary, this means avoid negative bias on secondary max 250V on primary and keep phase even a cheap 126C for 10mA max can give very good result

anyway you cannot affirm: "With IT coupling the frequency response and transient response get worse and hum/interference pick-up will increase."

but you can write: "in my experience i get bad frequency response etc"
 
Good quality interstage transformers, ratio 1:1, have wide bandwidth (depends also on source impedance, but over 100 kHz is no exception). This bandwidth is superior to the bandwidth of the output transformer, so it is not a restriction at all in a well designed amplifier.

Grids of output tubes like to see a low resistance path to mother earth, and that's a major benefit of interstage transformer coupling over capacitor - grid resistor coupling. One can discuss differences in sound, but that's done a million times already.
 
anyway you cannot affirm: "With IT coupling the frequency response and transient response get worse and hum/interference pick-up will increase."

but you can write: "in my experience i get bad frequency response etc"

It is not a question of my experience, it is a well know fact. Please show us one design where IT improves THD, frequency response and/or transient response.

In general, any transformer in the signal path degrade the performance compared to design without.
So why to add such components on purpose ?
Or is this intention to do nostalgic designs ?
(or is the answer the typical: sounds better...)

As a reference, Hammond 126A, look how it performs at 20 Hz driven with typical medium mu-triode or what sort of response it makes to 10 kHz square wave.

Hammond as a manufacture claims only this about 126-series IT's:
These units replace coupling capacitor circuits for improved sound.

If it improved the performance in any way, why would Hammond not mention about this ?
 
with IT coupling the frequency response and transient response improve
Compared to what? The bandwidth extremes of an interstage transformer are nowhere near as good as a coupling cap or direct coupling. While I do agree that 100kHz could be reached on a parallel feed interstage transformer, gapped ones aren't going to do as well (maybe 30kHz up top, 30Hz down low unless you spend big money). This isn't necessarily the end of the world on a zero feedback amp with so-so output transformers, but if you do end up needing to use feedback or your OP iron has excellent bandwidth, the IT is going to be a bit of a bottleneck.

One way to get the low DC impedance path to ground for the grid of the output tube is to use a grid choke.

and hum/interference pick-up decrease
No, IT iron is certainly sensitive to hum pickup. Cap coupling and direct coupling are not.
 
Last edited:
The bandwidth extremes of an interstage transformer are nowhere near as good as a coupling cap or direct coupling.

Which is of no consequence at all in a good design! Are you listening with your ears or your oscilloscope?
I already mentioned the advantage of inductive coupling with low resistance path ("blocking distortion" anyone??)

No, IT iron is certainly sensitive to hum pickup. Cap coupling and direct coupling are not.

No issue in a good design and layout.
 
Which is of no consequence at all in a good design!
If the IT transformer is going to leave the amp down 3dB at 30Hz, and the OT is maybe down 1-2dB at 30Hz, I'd say the design isn't "good".

It's totally fine to love using interstage transformers, but saying that they have great bandwidth and aren't another variable for magnetic noise coupling is just perpetuating ignorance.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.