Realistic mpa-20

[Psc]

Hey gang, I'm looking for documentation for a
Realistic mpa-20 PA amplifier. It's not the discrete version, it's the one with two kia7217ap chips. I'm looking to drastically modify it into a guitar amp. I'm hoping to leave the power amp section mostly the same, possibly tweak a few things and add a new preamp section.

The board is labelled P-200902.

I've only found a snippet that someone traced of the preamp section. Hoping I can find something a bit more comprehensive.

Thanks for your help!

[Kaz Kylheku]

According to the "representative circuit" and "application and test circuit" in the schematic for that KIA7217AP chip, it should be easy to convey current feedback straight to pin 7. For guitar, that's a great payoff for a minimum investment in parts and effort.

[Psc]

That's interesting. I'm still learning about this stuff, so correct me if I'm wrong here. Current feedback would mean taking the output, through a resistor to pin 7.

I've heard of this before but I'm still not clear as to why it's beneficial in guitar amps. Do you know of any good explainers on this? I'll keep searching.

The MPA-20 uses two of these chips and then goes through an output transformer. Is this necessary or beneficial? If I'm reading the datasheet correctly it is based on a 4 ohm load. Would it be fine at 8 ohm as well?

Thanks!

[Psc]

I just reread my last post and clearly I was trying to get out of work on time. The two chips are bridged and apparently outputting 20 watts. I'm not sure what that looks like because I don't have a schematic and haven't had time to trace it out.

Is the output transformer useful to me or would I be better off without it? It has taps for 4 ohm to 16 if I recall. My original plan was to have one output and then use a jumper for the output impedance based on the transformer. I'm very open to ideas.

Also, if anyone has any info about how one would bridge an opamp like the one I mentioned above I'd love to read it. I've done some searching but haven't been super successful yet.

Thanks!

[phatt]

Re Current Feedback also called "Defined Impedance" of an amplifiers output.
Most SS HiFi DC Poweramps have a very high Damping factor. No over shoot of the Cone, The speaker follows the signal as perfectly as the laws of physics allow.

Ye old Valve Guitar amps tend to have low damping factor. The cone over shoots more. Partly due to the Output Transformer. Also remember every time the speaker cone moves it also generates it's own energy Called Back EMF. High damping reduces that effect.

CF is no magic bullet yes at high volume you may notice a little benefit not so much at low volume. If this Amp has an output TX then you already have the best setup and little benefit will be had by implementing CF. (and likely more complicated because of the OTx)

Rod Elliott has lots of mind bending info if you wish to know more
https://sound-au.com/articles/cfb-vs-vfb.htm

Home page here; https://sound-au.com/

Now Others here will know far more about this interesting subject and Some claim great benefit but for me having messed with this stuff years back there is only  marginal benefit and I doubt if you are loud and live on stage you would hear much difference.

There are tricks you can implement with CF that can alter the Frequency response, may be of benefit,  Peavey implement CF in a lot of their circuits as do many others.

Warning As With any TX coupled output make sure you have a load on the output before you crank up the volume.

If it was on my bench I'd be using a stand alone preamp or some pedals first as messing with the unit directly might reap little rewards. just  a thought?

Regards Bridging Amps,,, cricky just get a chip that delivers the Wattage you want and save all the messin around.
Member J,M,Fahey has designed and built likely 100s of SS guitar rigs and a lot of those use OTx. Good to see that he is still posting
He will have a lot to say about the advantage of using Tx coupled outputs on SS rigs.
Phil.

[Kaz Kylheku]

Wow, an output transformer in a solid-state amp (let alone chip amp) is quite uncommon, I think. It has some advantages, like the multiple taps for different speaker impedances, a common feature of tube amps.

Output trafos in solid-state amps were common in the 1960's. There were reasons for that:

1. (I think): lack of output transistors capable of handling higher VCE voltages. Solution: change voltage with transformer.

2. Lack of complementary devices (matching PNP/NPN transistors). The output-trafo based designs followed the tube topology: phase inverter (transistor-based, of course), feeding separate identical output stages, both based on a NPN with a positive supply, but then feeding opposite windings in the trafo so that one pushes and the other pulls.

3. In regard to point (2), tube designers just doing what they know, but using transistors. Like when Fortran programmers continued writing Fortran in any language.

About current feedback with an output trafo, I have a hunch that using the signal from the speaker return after the trafo could introduce a phase shift into the feedback loop causing instability? The safe thing would be to sense the current on the primary side of the trafo.

Current feedback is often conveyed via not only a resistor but also a DC blocking capacitor. The reference circuit in the datasheet for these chips shows the external feedback resistor being AC-grounded via a capacitor. So the datasheet is calling for the amp to have unity gain at DC. Even if that Realistic amp happened to do something different, it's best to deliver the current feedback with DC blocked (with a large enough cap that you don't choke off the desired effects below 200 Hz). This way the circuit doesn't change any resistance seen by DC, and not upset any DC balance. (You wouldn't want the amp to have a DC offset, sending current through the transformer even when there is no signal.)

It's useful to have a switch which interrupts that feedback, so you can toggle it on and off. A simple SPDT on-on switch can connect the feedback in one position, and in the other position short out the current-sensing resistor so the amp is completely stock.

[Psc]

Hey phatt, thanks for the reply.

I'll definitely read the links you posted. I ran across that site in my searching but I wasn't entirely following it. Part of my issue is that I don't know exactly what I'm working with so I can't follow along, but I am keen to learn. I recently built a noisy cricket and really want to get more into solid state amps.

Your explanation of damping speakers and why in a guitar amp you would want more movement and back emf makes sense to me, and I get how you would get that with current feedback and also with an output transformer so that clears some stuff up for me.

I don't disagree about leaving the pa amp stock. The amp has four inputs, two microphone, an aux and a phono. My plan was to remove the phono, aux and one of the mic inputs, the one with the microphone transformer, and leave the second mic input. I want to build a preamp as a learning tool and also to take advantage of the already built power supply. I haven't decided on a preamp yet, but I like the look of the dirt road special.

My question about bridging amps is more to do with the fact that this amp is already doing that, so I'm interested in learning what's going on in there and if there's ways to modify/tweak. I totally agree that if I was to build from the ground up I'd just pick something else, or if I was to rob this for the parts maybe make two 5w amps rather than one 20w.

Thanks again for the information, it really helped me.

[J M Fahey]

This is a mid power (around 6W RMS into 4 ohm) car radio type amplifier.
You *can* turn it into a current feedback one but I can't read the fuzzy datasheet for instructions 
If available, post the PDF here.
I mean the real PDF,not a link to it.

PS: if bridged, forget about adding current feedback.

[Psc]

Kaz Kylheku,

All of that makes sense to me. Really good point on a DC blocking capacitor in the cf loop. And I'm sure you're right about wanting the cf on the primary side. You'd want it directly off the output I would imagine.

Do you feel like there's value in having current feedback as well as an output transformer? I'm sure there's no harm in trying it, just wondering about theory.

It is really interesting to see how older "traditional" technologies influenced the ones that came after them in ways that might not actually make much sense, but also using what you have to overcome the limitations of the new technology.

[phatt]

Some helpful info here: https://ozvalveamps.org/varidamp.html
variable damping been known way back.

Front page; https://ozvalveamps.org/
Site setup by a wonderful Aussie Roly Roper, sadly no longer with us but His site is a credit to him as he lists most of all the Aussie Valve amps that were made in Au and NZ.
With tons of good advice from a highly experienced Teck.
Phil.

[Kaz Kylheku]

Do you feel like there's value in having current feedback as well as an output transformer? I'm sure there's no harm in trying it, just wondering about theory.

This is all just guesswork about a unit without any schematics.

Speaking of which, by the way, Elektrotanya website has schematics for the MPA-40 which seems to have similar features. It has a discrete power amp, not chips, but also feeding an output transformer with multiple taps. From the schematic it is obvious that what is before the transformer is a voltage amp.

As far as theory goes, an ideal transformer would have no impedance of its own, and just serve as an impedance scaling device. If we assume it's driven by a voltage amp with zero impedance, then the speaker will still see zero impedance through the ideal transformer.

A real transformer has a lot of wire and windings, and a core. There is a DC resistance there and inductance also. A speaker that is behind an output transformer cannot see a near zero impedance, even if the amp provides that.

It's just a question of how much.

BTW this MPA-40 seems cool; I vant it. 

[dmeek]

Here is the schematic

You cannot view this attachment.

[J M Fahey]

Ok, ONLY NOW can we see the schematic.

1} FORGET current feedback and any fancy stuff.

2) *do* use the output transformer, you *need* it.
Output is bridged and IT primary winding combines out of phase signals.

So connect speaker to proper 4 or 8 ohm tap as needed.

PA amps are weird,this one has 70V taps and also is *designed* to be fed from a 12.6V car battery for street/field use.

[Psc]

Oh man, thanks so much for the schematic. This is so helpful.

J M Fahey, great to know about about the output transformer, I'll heed your words.

I haven't had a ton of time, but without the schematic I thought I'd at least do what I could, so I removed the one mic transformer and the inputs for the aux and phono inputs and installed a 1/4" speaker out jack. It has those spring clip speaker terminals, one of which was for the 70v that you mentioned. I couldn't see needing to use it so I unsoldered it from the connector, used it as my speaker out. Now all I have to do is jumper from that terminal to the impedance I need. So far so good.

I tried it out into a 2x12 at 4 ohms, and man, it's loud. It sounds quite good really. The tone control doesn't do much, probably not designed for guitar which I expected. I put a preamp pedal in front of it and it helped shape the tone quite a bit. It gets pretty fizzy at high volume. I'm guessing the sound of the chipamp clipping isn't the greatest. An assumption as I hadn't seen the schematic.

I also wonder if the mic inputs are just too sensitive, and maybe I should be tying into where the phono/aux does. Might tame it a bit? I'll have time to go over the schematic tomorrow.

Thanks everyone for the help!

[Psc]

I'll start a new thread, but any chance anyone has the schematic for the mpa-25? I have one of them as well.