The old impedance matching & output capacitor value questions

[OldManMatt]

Hello to anyone that reads this, hope life is being good to you. I have a question that seems to have been asked a thousand times yet I still need a little clarification. I hope I posted this in the correct area, and all though I'm not a newbie or beginner to audio I felt maybe these are beginner questions.

The last year or so I've been dabbling in DIY type stuff, replacing old parts on old gear, tube preamp kits, a couple dynakit rebuilds. So my question comes from me working on a tube preamp, fixing a couple issues that the prior owner left for me. I do have to take out the output capacitors on this preamp and I've been told I need to raise the value to make sure my bass rolloff is high enough for the low input impedance of the power amp..so here's my questions.

If the builder of the preamp designed the preamp to send the full bandwidth with say a .22uf output cap and there is no impedance mismatch between the Pre and Power amp, why do I need to raise the value? The power amp has a low 10k ohm input impedance. The output Z of the pre is like 100 ohms. Should be a big enough gap to cause no issues, but that depends on who you ask. If I follow the normal guidelines I read everywhere I should up the value on the cap to at least 2.2uf I believe, I didn't check the capacitor calculator but this number is what was thrown in my ear.
I feel like I understand how that works or the theory behind it, but if a preamp was designed to work a certain way with certain values and there is no mismatch do I need to follow these Capacitor rules? I listened to the pre for a short bit before I pulled it out for repairs/mods and everything was quiet and the bass didn't seem rolled off to me. I have a mic and should have measured but I didn't think it was an issue at the time. It wasn't til after when others were saying I need to up the value on my capacitor.

It's not like everytime I've bought a new amp or pre in the past that I've had to switch up capacitors to make sure everything sounded right, it's just plug and play. But in the world of DIY or "modding" all of a sudden I need big capacitors or I will have high bass roll-off. So obviously I don't know what I'm missing here..or I wouldn't be asking this question. Should I always follow the output capacitor value rule of matching it according to the input impedance of the power amp? If a builder says his preamp design is linear to say 10hz, is this only true if you are using an amp similar to what he tested on?
Thanks for any insight on this..and for reading my long post. I appreciate it.

 

[DVDdoug]

The series capacitor and load impedance make a high pass filter. 0.22uF and 10K is 3dB down at 72Hz.

Usually it's a "DC blocking capacitor". DC is zero Hz so it can't pass through a high-pass filter.

There is often a capacitor on the input (where the load is known) and on the output where the load (power amp input impedance) can vary.

If a builder says his preamp design is linear to say 10hz, is this only true if you are using an amp similar to what he tested on?

I don't know what that means. There's no reason for it be "non-linear" above 10kHz unless maybe there's an audio transformer somewhere in the signal path. ...There's also no reason to build a preamp with 1950's technology.

Tube power amplifiers need transformers on the output but you can build a tube preamp without one. (Microphone preamps sometimes have a step-up transformer on the input because it's an easy way to get "free" voltage gain.)

 

[DonH56]

Here are the simulation results for your two test cases. As @DVDdoug said the low-frequency corner is pretty high with a 0.22 uF coupling capacitor.

fc ~ 1/(2*pi*R*C), using R = 10k (ignoring the small 100 ohm resistor) then with 0.22 uF fc = 72.3 Hz and with 2.2 uF fc = 7.23 Hz. If you want 10 Hz you need a larger capacitor than 0.22 uF. Personally I would stick in something like 10 uF (fc = 1.6 Hz).

Many tube amplifiers have much higher input impedance, more like 100 k-ohms, which would provide a 7.23 Hz corner with a 0.22 uF capacitor. A SS amp with 10 k-ohms input needs a bigger capacitor.

HTH - Don

 

[MaxwellsEq]

I feel like I understand how that works or the theory behind it, but if a preamp was designed to work a certain way with certain values and there is no mismatch do I need to follow these Capacitor rules

A preamplifier designer can not know what power amplifier the purchaser will use. The designer of this preamplifier either disliked low bass or assumed a very high input impedance of the following power amplifier.

There's very little ultra low bass on most of my LPs. And LP playback can have low frequency issues with warps etc. So if this preamplifier only ever played LPs and was only ever matched with a high impedance power amplifier, it would probably pass muster.

 

[restorer-john]

Bear also in mind, the larger the coupling/output capacitor you install, the longer the time constant for charging it to '0V'. Larger caps will make larger thumps or pops and cause more woofer movement in a DC coupled power amp.

A single rail preamp with a capacitively coupled output will make a decent thump. Increase the cap and it will be louder and longer lasting.

Do your calculations as per @DonH56 above, and allow at least an order (x10) below or above the desired F3 you are targeting to ensure distortion is kept at bay.

 

[OldManMatt]

Thanks guys. I truly appreciate everyone's input, all of it was good information. I apologize for the way I worded some things, I can see how a couple things didn't make sense but I feel you guys were able to decipher it and give me my answer.

A lot of what I wrote I was just trying to use examples, and I should have just went straight to how and why my confusion and lack of confidence started. About 3 months ago I started on a tube preamp kit. I conversed with the builder before hand to let him know my purpose of the pre, which was to use in a digital streaming set up with a SS amp. I wanted to make sure the pre didn't have too much gain and there wouldn't be a problem with impedance matching between the 2. The SS did have an input impedance of 10k, the builder said at the highest point his pre was around 500ohms output.

As I was building it came time to put all my film caps in, the pre came with I believe was .22uf. I went online and used one of the cap calculators and felt I needed a cap pretty big in size compared to what was provided. I had some 8uf caps and really didn't want to spend money so I reached out to the builder and said hey, will using the ones I already have with this high of a value cause any issues with your design. His reply was I don't need to change anything up, what he gave me would provide a corner down below 10hz. This is where everything kinda went upside down for me.

So what he said didn't make sense compared to what I've read. But he is the builder and has been doing this a long time I've been told, this isn't something from across sea's. I didn't reply to him right away but it was bugging me so I replied the next day, tiptoeing around the "Are you sure" question and leaving my amp specs again. He says I'm good, the impedance match between the 2 components should be fine, im putting too much into this, people always want to change things, blah blah, and that if I insist I need to change it just double the .22 value. I got the sense he was irritated by me.

 

[OldManMatt]

Everything y'all gave me is solid information, I truly appreciate getting input from a few different people yet all of you are in agreement with each other. Thanks for sharing your knowledge and time.

 

[OldManMatt]

Oh yeah, so I ended up finishing the build with the low value caps provided, but when I hooked into the solid state amp I had ground loops, never could get a handle on it so I put a jdi duplex DI box between them just to get a taste of the pre. Everything sounded fine but the DI has I believe it has a 140K ohm input and 600ohm balanced output (my amp has both XLR/RCA) I didn't want to forever run thru a DI box so I used the pre on a tube amp I had for few weeks then got rid of the pre. Started a new project and this is why I came here for clarification. Thank you again

 

[egellings]

Hello to anyone that reads this, hope life is being good to you. I have a question that seems to have been asked a thousand times yet I still need a little clarification. I hope I posted this in the correct area, and all though I'm not a newbie or beginner to audio I felt maybe these are beginner questions.

The last year or so I've been dabbling in DIY type stuff, replacing old parts on old gear, tube preamp kits, a couple dynakit rebuilds. So my question comes from me working on a tube preamp, fixing a couple issues that the prior owner left for me. I do have to take out the output capacitors on this preamp and I've been told I need to raise the value to make sure my bass rolloff is high enough for the low input impedance of the power amp..so here's my questions.

If the builder of the preamp designed the preamp to send the full bandwidth with say a .22uf output cap and there is no impedance mismatch between the Pre and Power amp, why do I need to raise the value? The power amp has a low 10k ohm input impedance. The output Z of the pre is like 100 ohms. Should be a big enough gap to cause no issues, but that depends on who you ask. If I follow the normal guidelines I read everywhere I should up the value on the cap to at least 2.2uf I believe, I didn't check the capacitor calculator but this number is what was thrown in my ear.
I feel like I understand how that works or the theory behind it, but if a preamp was designed to work a certain way with certain values and there is no mismatch do I need to follow these Capacitor rules? I listened to the pre for a short bit before I pulled it out for repairs/mods and everything was quiet and the bass didn't seem rolled off to me. I have a mic and should have measured but I didn't think it was an issue at the time. It wasn't til after when others were saying I need to up the value on my capacitor.

It's not like everytime I've bought a new amp or pre in the past that I've had to switch up capacitors to make sure everything sounded right, it's just plug and play. But in the world of DIY or "modding" all of a sudden I need big capacitors or I will have high bass roll-off. So obviously I don't know what I'm missing here..or I wouldn't be asking this question. Should I always follow the output capacitor value rule of matching it according to the input impedance of the power amp? If a builder says his preamp design is linear to say 10hz, is this only true if you are using an amp similar to what he tested on?
Thanks for any insight on this..and for reading my long post. I appreciate it.

With audio preamps and power amps, impedance matching is of no concern when interconnecting them. Just don't let the power amp load the pre down to the point of distortion. Basically, you have a voltage source with a little bit of output R driving an easy load. Rule of thumb is power amp input R is =>10x preamp output impedance. Tube preamps are especially sensitive to too much loading, although tube power amps tend to have high input R's.

 

[DonH56]

With audio preamps and power amps, impedance matching is of no concern when interconnecting them. Just don't let the power amp load the pre down to the point of distortion. Basically, you have a voltage source with a little bit of output R driving an easy load. Rule of thumb is power amp input R is =>10x preamp output impedance. Tube preamps are especially sensitive to too much loading, although tube power amps tend to have high input R's.

The OP's issue was not really impedance matching, but rather choosing the right value for the coupling capacitor to reduce LF roll-off. See previous posts.

 

[Cbdb2]

Here are the simulation results for your two test cases. As @DVDdoug said the low-frequency corner is pretty high with a 0.22 uF coupling capacitor.

fc ~ 1/(2*pi*R*C), using R = 10k (ignoring the small 100 ohm resistor) then with 0.22 uF fc = 72.3 Hz and with 2.2 uF fc = 7.23 Hz. If you want 10 Hz you need a larger capacitor than 0.22 uF. Personally I would stick in something like 10 uF (fc = 1.6 Hz).

Many tube amplifiers have much higher input impedance, more like 100 k-ohms, which would provide a 7.23 Hz corner with a 0.22 uF capacitor. A SS amp with 10 k-ohms input needs a bigger capacitor.

HTH - Don

View attachment 300696
View attachment 300697

Would have been nice to see the difference in phase shifts.
As mentioned, the larger the coupling cap the lower its distortion ( less AC voltage across the dialectric ) Only a problem with electrolytic.

 

[DonH56]

Would have been nice to see the difference in phase shifts.
As mentioned, the larger the coupling cap the lower its distortion ( less AC voltage across the dialectric ) Only a problem with electrolytic.

This place is hell on free time. Ceramic (and some other) capacitors also change value with voltage. With a tube amp the distortion is rarely an issue since the DC bias is so large.