D3a in triode mode
#16 Re: D3a in triode mode
If you want I can run up a d3a using uTracer, I could run g2 and the anode at the same voltages for various g1 values and see what you get for Va and Vg2. But I suspect you have gone past that point now anyway.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
#17 Re: D3a in triode mode
That's a kind offer Nick thank you. I will carry on with my own experimentation but if you want to post a utracer result that would be useful. I really should get around to building the utracer kit I have on my shelf.
Once I'd figured out my labelling error I recorded the following.
Cathode 57R, Vak 124V, Ia 22mA, Ig2 6mA, Ik 28mA
I'll record more over the next few days fitting in with work.
Once I'd figured out my labelling error I recorded the following.
Cathode 57R, Vak 124V, Ia 22mA, Ig2 6mA, Ik 28mA
I'll record more over the next few days fitting in with work.
- IslandPink
- Amstrad Tower of Power
- Posts: 10041
- Joined: Tue May 29, 2007 7:01 pm
- Location: Denbigh, N.Wales
#18 Re: D3a in triode mode
Those numbers look very plausible.
"Once you find out ... the Circumstances ; then you can go out"
#19 Re: D3a in triode mode
Hmmm ... Didn't match my expectations from the curves. I was expecting 160Vak with 24mA total. 1.5v bias.
#20 Re: D3a in triode mode
Have I misread the chart from the datasheet ? 40V difference at anode seems very significant to me, not just rounding.
I designed the PSU to deliver 160V across Anode/G2 to Cathode, with 24mA through the PSU and Anode choke. the extra 4 mA means more drop through the PSU and Anode choke, but I'd expect that to self-correct to some extent, with reduced anode voltage leading to reduced current and settling more or less at planned operating point. When I've done this before it's worked out that way.
If I rework the design planning for 124V Vak, 28mA it all works out and the voltages at LC1 and LC2 outputs are as observed (within a volt or two). I'm just a bit disconcerted that the valve OP is way off what I expected.
I designed the PSU to deliver 160V across Anode/G2 to Cathode, with 24mA through the PSU and Anode choke. the extra 4 mA means more drop through the PSU and Anode choke, but I'd expect that to self-correct to some extent, with reduced anode voltage leading to reduced current and settling more or less at planned operating point. When I've done this before it's worked out that way.
If I rework the design planning for 124V Vak, 28mA it all works out and the voltages at LC1 and LC2 outputs are as observed (within a volt or two). I'm just a bit disconcerted that the valve OP is way off what I expected.
#21 Re: D3a in triode mode
I would have expected a 75R cathode resistor for the red marked op point. 20ma and 160v on the anode.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
- IslandPink
- Amstrad Tower of Power
- Posts: 10041
- Joined: Tue May 29, 2007 7:01 pm
- Location: Denbigh, N.Wales
#22 Re: D3a in triode mode
On my phono I think I have 90R cathode resistor and about 13ma total current and 115V on anode. Does that make sense ?
"Once you find out ... the Circumstances ; then you can go out"
- Paul Barker
- Social Sevices have been notified
- Posts: 8991
- Joined: Mon May 21, 2007 9:42 pm
#24 Re: D3a in triode mode
I’m making similar silly mistakes nowadays. Need to check and recheck current in calc etc.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
- Paul Barker
- Social Sevices have been notified
- Posts: 8991
- Joined: Mon May 21, 2007 9:42 pm
#25 Re: D3a in triode mode
Sometimes I’ve forgotten something and a google search finds me writing about it 20 years ago and my younger self pointing it out and telling where to find it Stephies pages, RDH or Terman.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
- Paul Barker
- Social Sevices have been notified
- Posts: 8991
- Joined: Mon May 21, 2007 9:42 pm
#26 Re: D3a in triode mode
Same is happening with knowledge loss about boiler repairs but I get the answer from my younger self lol
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
#27 Re: D3a in triode mode
plink plink fizzPaul Barker wrote: ↑Wed Jun 30, 2021 4:23 pm Same is happening with knowledge loss about boiler repairs but I get the answer from my younger self lol
#28 Re: D3a in triode mode
Ok so 68R +10R on cathode.
Scope square wave output connected across g1(ground (1khz). Cap across cathode resistor.
With scope switched off I'm getting 158Vak with 22mA total anode + G2. Better.
Switch the scope on and it changes, 130V, 28mA. The output measures at 3.2Vpp though so it's pushing g1 into positive grid bias. I'll see if the output level can be controlled from the Siglent, or use the Picoscope generator. I'm expecting less than 0.02V on g1 when it's built... (Cartridge output is 0.4mV Koetsu into 40:1 SUT
Scope square wave output connected across g1(ground (1khz). Cap across cathode resistor.
With scope switched off I'm getting 158Vak with 22mA total anode + G2. Better.
Switch the scope on and it changes, 130V, 28mA. The output measures at 3.2Vpp though so it's pushing g1 into positive grid bias. I'll see if the output level can be controlled from the Siglent, or use the Picoscope generator. I'm expecting less than 0.02V on g1 when it's built... (Cartridge output is 0.4mV Koetsu into 40:1 SUT
#29 Re: D3a in triode mode
Thats a lot of step up for a 400uv cart. I would have thought 10:1 would be better. Also remember all things being equal 20kHz is 20db Up on the 1kHz signal (not that there is that much signal at 20kHz, but pops and clicks are to be considered.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
#30 Re: D3a in triode mode
Hmmm.
The SUTs are Hashimoto HM-3 and the primaries can be configured in series or parallel for 1:20 or 1:40.
The Koetsu Rosewood Signature specs show an internal impedance of 5 ohms and nominal output of 0.4mV. I've never been able to work out whether thats RMS or peak, but I'm assuming RMS. Koetsu recommend a loading of 30 ohms but that seems low to me at only 6 times the internal impedance. Nevertheless in the phono stage we've already built using the this into the HM-3 at 1:40 and with a 47K load that reflects back as a shade over 29 ohms and it's working well. No doubt it can be improved on and of course this being DIY I can play with both the load and the turns ratio to hear what sounds best to me.
I went for 1:40 to increase the voltage as much as possible obviously, but I admit I have worried about the ability of such a small signal to drive the load well and particularly the input capacitance on the input valve. Triodes will have Miller to overcome and part of the reason for picking the D3a to experiment with is I can try it with both Triode and Pentode, the latter should greatly reduce the capacitance I think. I checked izzywizzy's designs (Steve ?) and I see he preferred the D3a input as pentode.
If I go 1:40 I'll have a Vpeak of 0.4mV * 1.414 * 40 = 23mV ish on D3a g1. If I aim for a 50R loading on the cartridge that would imply an 80K load (1600 turns ratio). With luck I'll get 70ish amplification factor from the D3a triode so approx 1.6V into the passive RIAA. Allow for that to be dropped to 0.16V after RIAA, a pair of output valves (TBC, but maybe 4P1L) across an LL1660PP transformer to give a balanced output.
If I go 1:20 that reduces the input to the D3a to around 12mV and to get a 50R loading I'd need 20K g1 load on D3a grid. More current to drive the Miller capacitance though. But less into the RIAA and SNR could be a problem.
Once I have experimented with that I'll look at using the D3a as pentode and also using either BJT or MOSFET driver before the D3a. I'd like to get more voltage before the RIAA I think.
Anyway, brain dump of my thinking so far. As always, all inputs are welcomed.