DIY 2C34 AB2 Amplifier Build - Odd Problems. - UK Vintage Radio Repair and Restoration Discussion Forum

Diabolical Artificer · 8th Jun 2025, 8:20 pm

I've recently been building another hifi valve amp using 5w 2C34 dual triodes paralleled class AB2. I previously built a lower power version which went without a hitch, this version is different only in layout & instead of one 2C34 per channel, there's two.It's exhibiting some very odd behavior indeed which I haven't seen before.

I wound the OPT's myself with an intended Zpri of 4k (spec sheet is 8k for one 2C34) but screwed up by winding two extra turns per secondary layer (5 primary's 3 secondary's interleaved), this resulted in a Zpri of 3K4. I also wound the mains tfmr which is working fine in terms of voltage & current specs.

To cut it short there seem to be two main faults/issues - 1) HF high THD/malformed negative going sinewave & 2) With the screens of the mains tfmr grounded to chassis a charge builds up between chassis & ground bus. This causes the LTP phase splitter to go out of regulation/kills transistors in the LTP CCS. The CCS for the LTP should be adjustable, but isn't. Previous built amps with the exact same LTP/CCS were adjustable. This changes when the mains tfmr screens ARE NOT grounded to chassis, but not much, but I still can't connect the groundbus to chassis without more adverse consequences.

I've tried changing the CCS's several times, changed the layout, de-coupled first stage/V1 as compared to the schematic, adjusted the quiescent current on the cathode followers, tried various bias points on the LTP, IE 4.5v to 8.5v, changed bias of OP valves, increased HT for V1 & V2, changed the load on the OPT - 8r & 18r & dis-connected the mains tfmr screen/EMF shield & finally earthed the chassis to mains earth.

Here's a video - https://www.youtube.com/watch?v=Qhk0l-yUfpg which hopefully shows the issues in more detail. See also attached schematic & pics. Any ideas most welcome, TFL, Andy.

G6Tanuki · 9th Jun 2025, 4:27 am

Given that the valves you are using were designed for use as VHF power oscillators and amplifiers, I would be looking closely at whether they are achieving their ambition to oscillate ( remember, every amplifier is an oscillator that just hasn't worked out how to oscillate yet!).

Adding ferrite beads to the grid and anode leads, along with stopper resistors of a few hundred ohms in the grid connections, and maybe anti-parasitic chokes - the usual ones being made from about 20 turns of enamelled wire wound over the outside of a chunky carbon resistor of a few hundred ohms.

Remember that a few inches of wire coupled with the input capacitance of a valve forms a VHF/UHF tuned circuit...

You seem to have strapped the grids of each double triode together - try separating them and then parallel feeding each grid through a 1k ohm resistor with ferrite beads on the resistor leads.

Think of each valve as being a push-pull VHF/UHF oscillator with the tuned circuits consisting of the connecting leads - if it oscillates at a hundred MHz or so there could be quite a bit of energy flying around to eat the CCS. Remember that the oscillation may only occur when the amplifier is actually being driven with the wanted signal, and then not at all points in the audio signal's waveform.

Debugging these sort of issues can be fun. How high in frequency does your scope go...??

Diabolical Artificer · 9th Jun 2025, 12:03 pm

I popped a couple of RF chokes on the anodes - 20uH/3A which I think has tamed the HF instability a tad. I my experience oscillation in valve amps is usually around the 30 to 50khz range, if higher even with a 10mhz scope ( which is what my scopes are) you can see HF oscillation even if you can't trigger & display the waveform.

Today it's acting differently - low OP on one anode of the LTP, could have sworn it was ok yesterday.

Ho hum, Andy.

Diabolical Artificer · 11th Jun 2025, 11:17 am

I'm currently working on the LTP as it isn't biasing properly so AC balance is way out of whack. I've triple checked connections & pinouts, poked it with a stick but can't find anything wrong. I have the CCS on the bench, it tests fine, put it in the amp, it doesn't. The only thing left I can think of is a dodgy valve base but they're NOS Cinch, very good quality.

The issue with tfmr I've put on the side as it were, solve one issue at a time. Andy.

8th Jun 2025, 8:20 pm	#1
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 8,409	DIY 2C34 AB2 Amplifier Build - Odd Problems. I've recently been building another hifi valve amp using 5w 2C34 dual triodes paralleled class AB2. I previously built a lower power version which went without a hitch, this version is different only in layout & instead of one 2C34 per channel, there's two.It's exhibiting some very odd behavior indeed which I haven't seen before. I wound the OPT's myself with an intended Zpri of 4k (spec sheet is 8k for one 2C34) but screwed up by winding two extra turns per secondary layer (5 primary's 3 secondary's interleaved), this resulted in a Zpri of 3K4. I also wound the mains tfmr which is working fine in terms of voltage & current specs. To cut it short there seem to be two main faults/issues - 1) HF high THD/malformed negative going sinewave & 2) With the screens of the mains tfmr grounded to chassis a charge builds up between chassis & ground bus. This causes the LTP phase splitter to go out of regulation/kills transistors in the LTP CCS. The CCS for the LTP should be adjustable, but isn't. Previous built amps with the exact same LTP/CCS were adjustable. This changes when the mains tfmr screens ARE NOT grounded to chassis, but not much, but I still can't connect the groundbus to chassis without more adverse consequences. I've tried changing the CCS's several times, changed the layout, de-coupled first stage/V1 as compared to the schematic, adjusted the quiescent current on the cathode followers, tried various bias points on the LTP, IE 4.5v to 8.5v, changed bias of OP valves, increased HT for V1 & V2, changed the load on the OPT - 8r & 18r & dis-connected the mains tfmr screen/EMF shield & finally earthed the chassis to mains earth. Here's a video - https://www.youtube.com/watch?v=Qhk0l-yUfpg which hopefully shows the issues in more detail. See also attached schematic & pics. Any ideas most welcome, TFL, Andy. Attached Thumbnails __________________ Curiosity hasn't killed this cat...so far.

9th Jun 2025, 4:27 am	#2
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 15,756	Re: DIY 2C34 AB2 Amplifier Build - Odd Problems. Given that the valves you are using were designed for use as VHF power oscillators and amplifiers, I would be looking closely at whether they are achieving their ambition to oscillate ( remember, every amplifier is an oscillator that just hasn't worked out how to oscillate yet!). Adding ferrite beads to the grid and anode leads, along with stopper resistors of a few hundred ohms in the grid connections, and maybe anti-parasitic chokes - the usual ones being made from about 20 turns of enamelled wire wound over the outside of a chunky carbon resistor of a few hundred ohms. Remember that a few inches of wire coupled with the input capacitance of a valve forms a VHF/UHF tuned circuit... You seem to have strapped the grids of each double triode together - try separating them and then parallel feeding each grid through a 1k ohm resistor with ferrite beads on the resistor leads. Think of each valve as being a push-pull VHF/UHF oscillator with the tuned circuits consisting of the connecting leads - if it oscillates at a hundred MHz or so there could be quite a bit of energy flying around to eat the CCS. Remember that the oscillation may only occur when the amplifier is actually being driven with the wanted signal, and then not at all points in the audio signal's waveform. Debugging these sort of issues can be fun. How high in frequency does your scope go...?? __________________ "It's not true that I had nothing on. I had the radio on!" -Marilyn Monroe . Last edited by G6Tanuki; 9th Jun 2025 at 4:34 am.

9th Jun 2025, 12:03 pm	#3
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 8,409	Re: DIY 2C34 AB2 Amplifier Build - Odd Problems. I popped a couple of RF chokes on the anodes - 20uH/3A which I think has tamed the HF instability a tad. I my experience oscillation in valve amps is usually around the 30 to 50khz range, if higher even with a 10mhz scope ( which is what my scopes are) you can see HF oscillation even if you can't trigger & display the waveform. Today it's acting differently - low OP on one anode of the LTP, could have sworn it was ok yesterday. Ho hum, Andy. __________________ Curiosity hasn't killed this cat...so far.

11th Jun 2025, 11:17 am	#4
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 8,409	Re: DIY 2C34 AB2 Amplifier Build - Odd Problems. I'm currently working on the LTP as it isn't biasing properly so AC balance is way out of whack. I've triple checked connections & pinouts, poked it with a stick but can't find anything wrong. I have the CCS on the bench, it tests fine, put it in the amp, it doesn't. The only thing left I can think of is a dodgy valve base but they're NOS Cinch, very good quality. The issue with tfmr I've put on the side as it were, solve one issue at a time. Andy. __________________ Curiosity hasn't killed this cat...so far.