Upgrading Cornet output capacitors

Braden · « **on:** 27 Sep 2003, 12:07 am »

I was wondering why the standard capacitor for C6 were 0.47uf 630V, and the recommended upgraded values are 1uf 200V? I used Jensen copper 0.47uf 630V as the capacitors for C6, should I have purchased the other values? What difference do those values make? Thanks!

hagtech · « **Reply #1 on:** 27 Sep 2003, 05:01 am »

Good question. I prefer a 1uF for output coupling capacitors because it can still deliver 16Hz rolloff into a 10k ohm load. Too small a capacitor means you lose bass, too big and you pass too much rumble. IMHO, 1uF is the best compromise considering the range of preamplifier loading that the capacitor will experience.

The ONLY reason I specify 0.47uF in the stock design is because there are no high voltage polypropylene type capacitors available from Mouser of greater value. That is the limit of that particular capacitor series. The upgrade path is to go with a 1uF auricap from AES. They cost a LOT more. I give you the choice.

My concept for a 1/2-kit is to keep the number of different vendors to a minimum. First, you buy the circuit board and intellectual property from me, then all other parts from other sources. I try to keep the "other" sources down to one, but in the case of a tube amp, I had to make it two - Mouser and AES (Antique Electronic Supply). Yes, I understand that it is more work for you the customer. However, it saves you a TON of money. You'd pay double if I had to kit it for you.

The cathode follower operates about 155V nominal, so a 200V capacitor is acceptable. Higher voltage ones have improved leakage.

jh

Braden · « **Reply #2 on:** 27 Sep 2003, 07:31 pm »

I guess I should have looked into it further before ordering all my parts. I guess the $40 :banghead: worth of jensen copper's could go elsewhere in the circuit! I did notice that I've had to turn up the level on my sub amps a wee bit. Still sounds good to me. Maybe I'll order them if I get anything else from parts connexion. I haven't played around with different brands, but would jensen caps be worth 3X as much as auricaps? Thanks again!

hagtech · « **Reply #3 on:** 28 Sep 2003, 06:33 am »

I haven't played around with the Jensens. Some guys swear by them. Is it worth 3x? Only you can answer that. Gotta go by ears. Same with tube choices.

Keep in mind, if your linestage has 20k ohm input impedance, you'll be fine with the 0.47uF. Even better into a 50k stage.

jh

GarMan · « **Reply #4 on:** 1 Oct 2003, 03:47 pm »

Jim, does this mean that if my integrated has an input impedence of 100K, the 0.47uF should be fine? Would a 1uF be too high for the 100K load?

hagtech · « **Reply #5 on:** 2 Oct 2003, 06:35 am »

Yes, with a 100k ohm load the 0.47uF will be fine. As a high pass filter, the low frequency cutoff of -3dB is given by:

f = 1/(6.28 * R * C)

or about 3 Hz. That will pass plenty good bass. It's also low enough to pass rumble, so your turntable needs to be quite good.

jh

PatOMalley · « **Reply #6 on:** 10 Sep 2009, 12:25 pm »

disregard.
-Meaning disregard my post. Posted in error.

tubesforever · « **Reply #7 on:** 10 Sep 2009, 04:36 pm »

Jim, would you verify my math so I can go on to make some general and interesting observations?

Just checking your formula in my Clarinet, it provides a 100k input resistance. So f = 1 / 6.28 x 100k x 1.0 uf.

I come up with a high pass crossover point of 6.28 hz. I am pretty sure this is correct.

If my math holds up, I can state for the record that I have zero woofer pumping at that crossover and luckily for me I don't hear any accoustic feedback at volumes I generally hit.

Before your Clarinet and Cornet 2 I did have accoustic feedback with the Precision Fidelity C7. Bruce Moore designs some fine gear, but the C7 exhibited woofer pumping as well.

I also had woofer pumping with the Bedini 6677 phono and line stage. I think John told me the circuit could deliver 1 to 100k bandwidth. I will say the bass on this phono stage was outstanding and the RIAA is about the best I have heard as well. However, overall the sonics and sound stage were congested and there is no SET magic with this solid state circuit.

My Audio Research PH3 phono stage sounded decent and had no woofer pumping but I never achieved the clarity or SET magic with it that defines your Cornet 2 phono stage. With the Russian teflon bypass caps, I get more dynamics as well.

Knowing what I understand today, I think your design provides the best opportunity to achieve excellent sonics with the least amount of parasitic losses from woofer pumping.

Kudo's Jim.

Brinkman · « **Reply #8 on:** 10 Sep 2009, 06:24 pm »

Tubes,

Using the Clarinet (100K) and a 1uF cap, the -3dB pole is:
f = 1/[(2xpi)(100000)(.000001)] = 1.59Hz

Increase the cap to 3.0uF and f = .53Hz. That's really close to DC. One might want the input cap of the Clarinet back in there.

hagtech · « **Reply #9 on:** 11 Sep 2009, 01:03 am »

In some [marginal] designs the woofer pumping can be due to power supply issues, inter-stage feedback, nonlinearity, clipping, RF oscillations, etc.

jh

jameshuls · « **Reply #10 on:** 15 Sep 2009, 04:41 pm »

How do you measure the input impedance of an integrated amp? Or do you just open it up and look for a resistor?

hagtech · « **Reply #11 on:** 18 Sep 2009, 01:10 am »

You could, but impedance can be a complicated thing, as it varies with frequency. Most amplifiers will have a 50k to 100k input resistance at low frequencies. Some Pass units go pretty low I think, perhaps 10k or so.

jh