Need help understanding AC187/AC188 radio amp circuit

[DF96]

Identifying the relevant resistances will not be straightforward. For the cap between the first and second stage you have to consider the output impedance of the first stage and the input impedance of the second stage.

The first stage output impedance will be a bit less than 3.3k, because the resistor has the first BJT collector in parallel with it and there is negative feedback via the 1M resistor. The actual impedance here will depend to some extent on the impedance of the signal source. It all gets complicated, so I would assume 3k and not worry too much.

The input impedance of the second stage is set by its current (9mA?) and its current gain. Then in parallel with that there is the 120k resistor. Wild guess say around 5k.

Then there is the 1k series resistor so you get a total of 9k to combine with the 640nF cap.

 

[bancika]

I didn't have 680nF caps, so I put 150nF for the first one and 470nF for the second one. Will test later today if it passes enough lower frequencies. I guess anything over 70hz would be good.

 

[Mooly]

DF96 explains it well. Everything interacts with everything else to put it simply.

I had the 3k3 as 33k (oops)

Do you want the simulation file to play with ? It will run straight off in LTspice. And here is the response with 3k3 instead of 33k.

 

[Mooly]

I didn't have 680nF caps, so I put 150nF for the first one and 470nF for the second one. Will test later today if it passes enough lower frequencies. I guess anything over 70hz would be good.

Here are those values.

 

[bancika]

I don't have LTSpice and don't know how to use it. It's on my todo list, but never have time to do it I even created software (DIYLC, you may have heard) for drawing layouts and schematic, just to avoid learning LTspice and similar. I find them very complicated

Can you please just try how it looks with 150nF input cap and 470nF for coupling? I promise I won't bug you anymore.

Cheers,
Bane

 

[Mooly]

The second image is for 150/470nf (refresh the page if its not showing as we posted together)

Click my signature line for installing and using LTspice.

Here is the file in case you get curious. Just click and it will run.

 

[bancika]

thank you, this is very helpful. I put the two plots on top of each other and stretched to match the scale. Looks like the roll-off starts around 300Hz in the second one and it's around 2db lower at 100Hz. I'll probably add some capacitance to flatten the response between 100Hz and 300Hz.

thanks again, I owe you a beer or whatever you like

 

[Mooly]

and your very welcome

Here is the original with 0.64uf and yours with 150/470nf. You can see where the response is 3db down from the 1kHz point.

 

[Mooly]

If you replace both with a couple of 4.7uf electrolytics then you are pretty much flat all the way down (different X axis scale).

 

[bancika]

Yes. Although, some bass roll-off is desirable here, otherwise it may sound farty or boomy when it gets overdriven by strong guitar signal. I'll try it tonight with my guitar speaker. It also has roll-off around 100Hz (plot is below). If there's not enough bass, I'll just add another capacitor in parallel with the existing ones. Btw, my output capacitor is 470uF instead of 400uF, so it may compensate for some of the bass loss

An externally hosted image should be here but it was not working when we last tested it.

Cheers

 

[Mooly]

Yes, you will just have to try it and see how it performs. Good luck

 

[bancika]

Reporting back, I tried two variations and I think there's slight difference in favor of higher capacitance and I stuck with 0.47uF in fist and 1uF in the second position. It sounds just awesome with my Strat guitar. Very clean and warm...and very similar to good tube amps. There's slight breakup when pushed hard, but it's very pleasant. And it's surprisingly loud. With my efficient guitar speaker it's much louder than the regular TV level. Even maybe too loud for apartment.

Thanks to both of you for your help.

 

[Mooly]

That's really they don't make 'em like that anymore.

 

[Ian Finch]

I haven't seen carbon composition resistors like that since the mid 1970's when the last plant here closed. You might want to check the actual values before assuming the markings are correct, since their drift could be in the hundreds of %

 

[bancika]

yes. The couple I checked are in the +10-20% range...but I won't do anything about it as it sounds really good
in some tube amps I've played with, the old carbon comp resistors look kind of dry and crumbly, but these look really good. I have some new production carbon comp resistors and they look about the same.