Hell-marriage amp (5842-> IT -> 2A3)

What people are working on at the moment
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#136

Post by Dave the bass »

Mike H wrote:How long do they take to start doing it from cold?
Mostly after about 20-30 mins. At our-Phils it did it almost straightaway but I think that was a noisy bias pot. It's fixed bias remember, pot has been replaced by fixed value resistors.

DTB
"The fat bourgeois and his doppelganger"
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#137

Post by Mike H »

Righty-ho.

Still don't have an answer!
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#138

Post by Dave the bass »

Kitchen table take-over again today.
Image

Went further into why the TJ 2A3 anode current increases randomly. Still can't find out why, grid voltage still steady, HT steady. Like for like voltages on each channel, tried Nicks suggestion of removing the fixed bias pot wiper caps from the equation, nope still the same. Checked cap values out of cct, both channels cleared.... Its only the valves themselves and the bases left! Odd.

So, I did what any fine technician would do, replace the TJ's and put them in a box and fit 2A3C's which don't cause the same conditions :)

Needing to lift the spirits a bit I thought I'd have a go at changing the front end around a bit. A few pages in from the start of this thread our-Will mentioned the Russian 6C45 high mu triode. I bought a quartet to play with and have now replaced the 5842 with a 6C45.

I'm running battery bias in the cathode, approx 1.3V with the anode at 150V and ~22mA. Seen a few different O.P.'s for 6C45 and seeing as I didn't have any 62R 2W resistors to hand I've settled on batt bias for now just to see how that works/sounds. Also fitted a 1K grid stopper after the chap on the VT52.com site mentioned he found his 6C45 oscillated when he was testing them.
Image

Tomorrow my attention will turn to 6EW7 amp and major fettle-edge.

I luv this hobby :-)

DTB
"The fat bourgeois and his doppelganger"
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#139

Post by Mike H »

Weird, the TJ's seem to have penchant for it.

Grid poisoning? Gas?
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#140

Post by Dave the bass »

Mike H wrote:Weird, the TJ's seem to have penchant for it.

Grid poisoning? Gas?
Dunno TBH Mike, they are knocking on a bit and have brought lots of musical pleasures to my ears over the years.

I'd be cautious of saying TJ's have a penchant for it though, it's only 1 pair (mine) out of many that seem to do it on one (mine!) amp possblee in all fairness.

DTB
"The fat bourgeois and his doppelganger"
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#141

Post by Dave the bass »

Breaking News from the cutting edge here in Durfud!

I remembered our-Nick and our-Mike mentioning resistors across the secondary of ITx's. I've been fettling in different values of resistance and its really interesting to see the effect on sinewaves at either end of the bandwidth. I had a 220K across the secondary and its didn't seem to alter (tidy?) the shape much but a 12K had a very dramatic effect at both ends of the spectrum.

Cor, the effect of the reflected resistive load back through the IT I'm guessing?

I noticed the 2A3 bias changed as I lowered the value of the resistor across the secondary too so I'll fettle that back to -45 and recheck. compare one channel to another unmodded.

DTB
"The fat bourgeois and his doppelganger"
Deco-Doctor
User
Posts: 177
Joined: Sun Oct 28, 2007 8:58 am

#142

Post by Deco-Doctor »

removed
Last edited by Deco-Doctor on Fri Apr 27, 2012 1:05 am, edited 1 time in total.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#143

Post by Mike H »

Revision to earlier statement ~ these particular TJ's seem to have a penchant for it :D

Could be they're getting old; gas is not just what may leak in from outside, it can also come off metals and stuff inside, hence the suggestion above ^ to reduce the operating temperature :?:

I am reminded of reading American net topics where guys are complaining about 300B's not lasting very long (so which makes to go for that last the longest and all that). Then the penny finally drops that they keep trying to run them at 32 Watts or something :D

Dave the bass wrote:Cor, the effect of the reflected resistive load back through the IT I'm guessing?
Yep.

I presume the valve using the IT is OK with a 12k load?

I noticed the 2A3 bias changed as I lowered the value of the resistor across the secondary too so I'll fettle that back to -45 and recheck. compare one channel to another unmodded.
That implies a current is flowing through the IT secondary and with resistor making a Voltage drop, wouldn't have thought it should be that much to make that sort of difference. Image
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#144

Post by thomas »

Hi DTB glad to see you're busy! I've been tweaking my amp a bit since your visit (un snubbing mostly..!) but have yet to try the resistor across the IT secondary...
I had a 220K across the secondary and its didn't seem to alter (tidy?) the shape much but a 12K had a very dramatic effect at both ends of the spectrum.
What did this do, exactly?!

I presume the valve using the IT is OK with a 12k load?
As a total learner I has to ask... what would this sort of relatively low value 'do'?!
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#145

Post by Mike H »

"Un-snubbing" eh? Image

What did this do, exactly?!
The resistance value is "reflected" (as we say :D ) back through the IT so that's what the valve "sees" as the resistance value of the load.

As a total learner I has to ask... what would this sort of relatively low value 'do'?!
Depends what the Ra (anode resistance) of the valve is, if it's high and the load is too small for it, it don't work too well. Basically it's over-loaded. In general terms you lose gain and the distortion goes up.

According to the experts here the load should, ideally, not be less than 3 times the Ra of the valve. Or is it 2 times?

And how do you find out what the Ra is, you may be entitled to ask? :D

The better more comprehensive datasheets should include a graph with a confusing number of obscure looking curves on it. The bottom ('X') axis is anode current (Ia).

One of the curves coming down from the top left corner area is Ra, there's a kilohms scale for it somewhere on the left margin, and it will be noted that it decreases as anode current increases (further to the right).
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
Mike H
Amstrad Tower of Power
Posts: 20157
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#146

Post by Mike H »

Appendix ~ effectively Ra is the resistance of the valve, so it's reasonable that if you increase the anode current but keep the Voltage the same, this resistance decreases.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#147

Post by Dave the bass »

Deco-Doctor wrote:Have you tried running them a bit cooler, ie backing off the current to below 50ma? Then see what happens.
Hi ya Martin, yep, tried that. They just seem to have a mind of their own.
Deco-Doctor wrote:As you say they are getting on a bit and may be slightly gassy and have a bit of grid emission. Does the anode current remain stable if you put them in a cathode biased amp?
I haven't got a cathode bias amp to hand at the mo so TBH I don't know the answer that yet.
Mike H wrote: I presume the valve using the IT is OK with a 12k load?
Yup, it's a chunky little beast the 6C45p, our AndrewL has a spud amp using one. He drove the massiff Sachiko (sp?) speakers with it once at an Eggy-bash.
Mike H wrote:That implies a current is flowing through the IT secondary and with resistor making a Voltage drop, wouldn't have thought it should be that much to make that sort of difference. Image
Yup, I'd have thought that too Mike but the effect on the waveform is very pronounced. Here's a 120Khz sinewave 'getting tidied up'... click pic for mini movie...
Image

It was just an experiment really to see 'what happens if....' but after re-jigging the (fixed) bias back correctly to -45 on the grid (after the introduction of the 12K resistor threw it right out) the amp sounds pretty good. I'd imagine it's not conventional to have a 12K grid leak pulling 3mA through it to ground on the -45 grid bias feed but the amp does seemed to have gained some 'grunt'. I'm guessing the 6C45p is seeing a 12K load though the 1:1 ITx?

Bias on the ShuGuang 2A3C's are steady, usual 250V @ 50mA or so OP, no drift using these.

The best part of a whole weekend 'playing' with 'scope and sig gen and amps.

DTB
"The fat bourgeois and his doppelganger"
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#148

Post by thomas »

Aha
Thanks for the tech.
Couldn't find one of those graphs online for DTB's 5842 but its 'typical operating conditions' Ra is apparently 1700 ohms, so the 12k would appear to be ok for it....
Don't want to hijack DTB's thread and stop me for doing so... however, I will ask-wot does the 12k resistor across its 'grid' do to the output valve? (shows ignorance here.. :? ) Various schematics I've seen suggest 220k or so would be a usual value here if an interstage tranny wasn't there...
User avatar
thomas
Old Hand
Posts: 910
Joined: Sat Sep 26, 2009 1:52 am
Location: N.W.Kent

#149

Post by thomas »

Oops posted after DTB's vid-! That sine wave certainly does 'improve'...Did your 5842 do that too, DTB?
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#150

Post by Dave the bass »

thomas wrote:Oops posted after DTB's vid-! That sine wave certainly does 'improve'...Did your 5842 do that too, DTB?
Yeah, but I didn't faff with the 5842 as much TBH. I just wanted to see if I could fettle in a 6C45p in it's place partly as a test of my new-found-confidence ( :-) ) and partly just to see what it sounded like after our-Will reported better results with one up front.
"The fat bourgeois and his doppelganger"
Post Reply