Hi magz, nashbap
If you're looking for a compact reference on designing for that resonance, Frank's electron tube data has a copy of Terman's 1943 Radio Engineer's Handbook here: electron Tube Data sheets - No brand related documents
Beginning on page 371 there is a discussion of "Miscellaneous Coupling Systems for Audio-frequency Voltage Amplifiers". The discussion of "Transformer Coupling with Shunt (Parallel) Feed" begins at the bottom of that page, with relevant graphs at Figure 14 (b) on page 374 illustrating the effect of altering the value of the coupling or "parafeed" capacitor.
Enjoy
I downloaded Terman's handbook and went thru the relevant pages. I experimented with the cap value and the effect is as represented in fig 14b.
Thank you for posting this info in the thread. It will come in useful.
nash
Based on your experience, I decided to experiment a bit.
I was getting a peak around 25hz of about 7db higher than the flat line somewhat like your blue line but not as severe a peak as yours. I have built according to Ale Moglia's LowGain model that has a 0.22uf coupling cap and a 510k resistor after it to G.
First, I paralleled 0.5uf with the 0.22uf and found that the peak remained at 25hz but it had dropped to about 3.5db and the curve portion less than 25hz was now sloping very gently upwards.
Next, I combined a 10k resistor to the 0.5uf and paralleled the combination to the 0.22uf cap. Wow, the peak dropped to only 2.5db and the the curve portion less than 25hz was now very gently sloping downwards. Also the peak had nicely rounded out. Next, I substituted a 20k for the 10k . Here I found that although the peak had remained about the same, the portion of the curve below 25hz was now sloping down, maybe a bit more than your green curve. What you have suggested most definitely works. Thank you!
I downloaded the zip file you attached, but am a bit confused as to how to use it. I entered 0.22 under C Parafeed Design2 keeping all else unchanged and Dr P optional Parts shows
0.44 0.44 uF est. 597149 8285 Ohms
Not sure how to interpret this.
nash
Last edited:
To use the calculator you need to fill in all the values in the Design 1 and Design 2 columns D and F. Columns D and F are all you need to enter, the rest are calculated by the spreadsheet. C_Parafeed is your small cap. C_DrP is your larger cap. R_DrP is the resistor in series with the large cap. The diagram on the right shows what the component names correspond to. Usually I put the plain parafeed as Design 1 and the lossy parafeed as Design 2 so I can see the differences.
The results appear in the lower section, Output Peaking, -3dB Frequency, etc. and in the graphs.
Just play with different C_Parafeed, C_DrP and R_DrP values to get the peaking reduction and bass extension you want.
The results appear in the lower section, Output Peaking, -3dB Frequency, etc. and in the graphs.
Just play with different C_Parafeed, C_DrP and R_DrP values to get the peaking reduction and bass extension you want.
Last edited:
I use a 1uF Duelund Copper Cast paralleled with 3.3uF Mundorf Supreme - 8080R TX2575 resistor. I'm not a huge fan of the Russian teflons, but I love the Duelunds.
Thanks for your responses.
I am wondering how much of a contribution to the overall sound is from the mu gyrator cap and how much from the output cap? Just speculating, perhaps that's why Ale suggests a PIO cap of 0.1uf for the gyrator and a 0.22uf Teflon for the output. I am sure you all have over the years played with the combination; if so please share your experiences.
Magz, were there notable sonic differences between before inserting the Dr P cap and resistor and after? As noted before I can see the measurements clearly improve with it but I have not yet implemented it, so asking.
Thanks. nash
I am wondering how much of a contribution to the overall sound is from the mu gyrator cap and how much from the output cap? Just speculating, perhaps that's why Ale suggests a PIO cap of 0.1uf for the gyrator and a 0.22uf Teflon for the output. I am sure you all have over the years played with the combination; if so please share your experiences.
Magz, were there notable sonic differences between before inserting the Dr P cap and resistor and after? As noted before I can see the measurements clearly improve with it but I have not yet implemented it, so asking.
Thanks. nash
From memory, without the RC the bass is boomier/ less controlled and the lowest frequencies are less distinct.
Yes I'm using them in the same build as You, they are excellent, I find the FT3's very good, but a little on the (over) analytical side, with the Miflex you get a lot of speed, dynamics & detail but without any hardness, you do get nice timbre too to the sound, they probably ain't going to be as good as Duelands, but from what people tell me they are similar, if you have a problem with heat inside your build you might want to try KFPM-1 from Miflex, they apparently sound as good but without the problems that can arise with oil caps.
So what value did you use in place of the 510k resistor?
I realise that you are asking for cap recommendations but figured to include the following for if/when you decide to 'keep going' and need some inspiration:
Replacing the CCS load (Ales Gyrator) with ISO RC-160-15W plate chokes, removing the coupling cap and then connecting 01A plate direct to grid of the next stage was the biggest improvement I'd had with these things in many years.
Last edited:
I'm another fan of plate chokes mostly because I have some nice amorphous ones. I use FT-2 teflon coupling caps in my 2-stage 2a3 amp (but driver isn't 01A). But if I went for a 3 stage solution, direct coupling an 01A to the second stage looks like a very nice idea. A 3 stage DHT amp is an alternative to a 2 stage one with an IDHT driver.
Ale's gyrator works best with His source follower, but as I haven't tried plate choke I can't comment on that, but I may well try it at some stage, Andy Evans might well chime in here, that more His bag too, when I went from plain gyrator to adding a source follower, there was also a big difference, another big difference was going from Rod Colemans V7 regs to the new V9's.