Solid State Guitar Amp Forum | DIY Guitar Amplifiers

Solid State Amplifiers => Amplifier Discussion => Topic started by: HamSandwich on November 29, 2015, 03:59:22 PM

Title: Variable Amplifier Impedance
Post by: HamSandwich on November 29, 2015, 03:59:22 PM
Hola - first post here!

I'm looking at chipamp poweramps and came across some posts about changing them from voltage amps to mixed voltage and current amps in order to make it a little more... forgiving, ala Project 56 (, Figure 2, at Rod Elliot's Sound Product site.

I won't pretend to understand a lot of what's going on here, but I've been looking at poweramps used in production models, such as the Marshall 8040 (, and trying to figure out why values were chosen.

I'm somewhat of a novice with LTSpice, so when I try to simulate figure 2 from ESP, I get the 18.8V output without the 8ohm load, but when I make the connection to an 8ohm resistor, I get a square wave at 200mV on the output of the amplifier. So before I can use the equations to figure out output impedance, I need to get that voltage under load somehow.

In any case, does anyone have any tips on how to figure out that voltage before I continue on the path of misunderstanding?

Also, I did what I would consider a pretty thorough search on chipamps set up like this for guitar amps, but I can't seem to find many comments any projects posted here, or elsewhere. Of course there are a ton of people using the datasheet examples of chipamps for power amps, but nothing similar to this. I thought that was a bit odd as prevalent as these chips are in current commercial amps. So if there is something similar going around, I'd love to know about it! Thank you. 
Title: Re: Variable Amplifier Impedance
Post by: Enzo on November 29, 2015, 05:18:36 PM
Not my deepest subject, but considering all the chip amp output stages in guitar amps I know, the vast majority seem to be simply data sheet circuits.   I am sure there are good examples of current feedback and such, they just aren't coming to my mind.  More complex output stages tend to be discretes.   I never thought of any of them as unforgiving.  Most chip amps have overcurrent protection.  There is little point in overdriving them, so your tone comes from the preamp.

I don't use Spice, so my input is minimal, but if the thing makes a good waveform into no load but collapses under load, my reaction is what it would be in a real circuit - it cannot provide the current the load demands.  So when this happens, what are your power rails doing?  How much current can your sim power supply provide to the circuit?
Title: Re: Variable Amplifier Impedance
Post by: HamSandwich on November 29, 2015, 05:58:26 PM
Hello Enzo, thanks for the reply. I have a handful of schematics I'll link to tomorrow for what I was looking at that use the mix mode feedback rather than a straight copy of the datasheet/gainclone.

I'll have a look at what the power supply is doing as well! Cheers!
Title: Re: Variable Amplifier Impedance
Post by: phatt on December 03, 2015, 05:07:18 AM
Peavey amps with Transtube have that system so try one out if you get a chance. :tu:
I messed around with it a long time back,,, but no magic bullet to be found.

Not really noticeable (to my ears) at low levels but higher volumes it may add a little extra.
Transformer coupling might deliver more rewards but by all means try the mixed mode feedback.
Title: Re: Variable Amplifier Impedance
Post by: Roly on December 03, 2015, 09:49:41 PM
I'm inclined to agree with Phil; the output impedance is only part of the story.  In short; a s.s. amp has a Zout of a fraction of an ohm while a valve amp might be around 5 ohm, implying the s.s. amp has an iron grip on the speaker cone and thus high damping factor, and the valve amp doesn't (i.e. more colouration by speaker and cab).  So it's more a matter of magnitude than precise value.

The B-H curve of the iron in any OPT is also a serious contributor  to "valve sound".

These topics can be discussed endlessly, but the only test that counts is what it sounds like (to your ear/taste), and the only way to know that is try it.   :dbtu: