K4020 stereo Mosfet Amplifier

Kits Audio Hi Fi Projects
1

02/26/2010

Hello,

I purchased one of your K4020 stereo Mosfet amplifier kits from Hobby Tron.

The serial number is #0903.

Although I once studied and then taught electronics up to about 35 years ago, I left the field and pursued another career. I apologize whole-heartedly if I have made mistakes regarding the following.

I have a number of questions.

The schematic on page H4020-ED2-55 shows T10 and T11, both of which are IRFP9140 P-channel Mosfet’s connected by what appears to be with the Source and Drain reversed. The Drain of each is connected to R78 and R79 (0.22 ohm) respectively, both of which are then connected to the +40 volt DC supply.

With Op Amps A1 and A2 driving the Mosfet’s Gate, Drain to Source and feedback for A1 and A2 being the voltage drop across each 0.22 ohm to Drain, I just cannot seem to see how this could possibly work.

But then after carefully studying some of the traces on the circuit board (P4020’2), the drains of all the Mosfets T10 & T11 (IRFP9140 P-channel) and T12 & T13 (IRFP140 N-channel) appear to be connected together. If each of the Op Amps drives the connected Mosfet Gate with each respective Mosfet Source connected to it’s respective supply voltage via the source resistors (0.22 ohm), the circuit seems to me it should then work.

Confused by what I found, I downloaded what appears to be some revision (manual_k4020.pdf), which confuses me even more.

Mosfets T10 & T11 are depicted the same as the printed manual that came with the kit.

The circuit board part number in the PDF is P4020A’3. The circuit board that came with the kit is P4020A’2.

The schematic’s have several differences though.

The kit has a 3.3K ohm resistor for R52 and no high frequency limiting capacitor across R59 a 39K ohm resistor on the negative feedback side of IC7 (TL061 Op Amp). This gives IC7 an apparent gain of 11.8 (Av = 11.8).

The PDF drawing shows a 33K ohm resistor for R59 and a high frequency limiting capacitor of 33pF across R59. IC7 in the PDF circuit has an apparent gain of about 1.18 (Av = 1.18).

With a gain of 1.18 (almost unity, no gain at all) and with an input sensitivity of 1 VRMS, resistors R53 & R54 (both 180 ohm) feeding the step-down series voltage regulator transistors T5 & T7 for IC7, will only see a maximum voltage swing of 1 Volt peak to peak to drive Mosfet Op Amp drivers A1 through A4 with.

This seems to me it will yield less than 10 Watts maximum into an 8 ohm speaker load, so something does not seem right.

C8 was increased to 330pF from 1nF originally which will roll off high frequency response of the amplifier much sooner. I don’t see a problem with this, but perhaps I am missing something in my understanding how this change works somehow in conjunction with changing R59 from 3.3K ohm to 33K ohm.

I found another change involving the protection circuit, primarily ZD5. This Zener is labeled and used as a 4.3 Volt Zener in conjunction with the ±18 VDC supply being OK (+V1 to –V1) applied to the base of Darlington T9, but then ZD5 is shown used again labeled as a 43 Volt Zener in series with ZD7, another 43 Volt Zener in conjunction with the ±50 VDC supply being OK (+V3 to –V3) applied to the base of T3.

In either case, if ZD5 is 43 Volts in the (+V1 to –V1) supply OK or if ZD5 is 4.3 Volts in the (+V3 to –V3) supply OK circuits, neither will not work unless I am really missing something here too.

So here are a few more questions.

Is there a problem with my circuit board?

If there is a problem with my circuit board what needs to be done to correct it?

Are there perhaps problems with both of the schematics?

If so, can you provide me one that is correct?

Also, is there an explanation of how this amplifier works? If the schematics are correct, I have no idea and I will probably do something incorrectly, rendering damage when I build it.

Please Help.

Sincerely,

Richard Borkowski
837 Lindley Road
Bronson, Michigan 49028
E-mail: Rborkowski@vbs.net

2

Sorry, no circuit descriptions.
It is not clear from your discription: does you amp work ?
Please be as brief as possible.

3

03/01/2010

Hello,

I am very surprised by your short reply to all my questions.

In reply to your question if I have built the amplifier yet.

No I have not nor would I dare to because I found major errors in the schematic diagram I wrote you about and I don’t want to cause damage to it unnecessarily.

I am now quite confident in my findings.

I spent some more time to better define the problems I have found, pictorially using both your schematic diagram, a portion of a diagram I drew to show how I believe it should be, descriptions of circuit breadboard tests I performed and the results I measured.

I apologize that I cannot be any more brief as you requested than herein.

Once again then in regards to the schematic on page H4020-ED2-55.

First problem.

Your schematic shows T10 and T11, both of which are IRFP9140 P-channel Mosfet’s connected by what appears to be with the Source and Drain reversed. The Drain of each is connected to R78 and R79 respectively, both of which are then connected to the +40 volt DC supply. Below is one segment of the schematic, showing T10 as I described it to be above.

Since I cannot paste the image here, I will email it to you.

I constructed this circuit on a breadboard. With any voltage plus or minus up to 10 volts applied to the gate, a current could be made to flow but only 0.7 volts maximum would conduct regardless. I conducted this test using 5 other P channel Mosfets of different brands. Same result in every case.

The reason for this is there is a parasitic diode (also often termed Body Diode) that is part of every Mosfet transistor device. This is an integral reverse p-n junction diode between the Drain and Source. Therefore with a positive voltage applied to the drain instead of the source of a P channel Mosfet, only the parasitic diode within the Mosfet will be able to conduct current.

Under normal circumstances it isn’t sensible to bias this diode within the Mosfet into conduction as it has a very long recovery time to cease conducting once signal is removed or even with the gate capacitance entirely discharged.

I redrew this same portion of your schematic to visually show how I believe the circuit was intended and I included the parasitic diode for the P channel Mosfet symbol. It should now be very clear to see and understand.

Since I cannot paste the image here, I will email it to you.

Next problem.

Your schematic shows a 33K ohm resistor for R59 and a 39K ohm resistor for R59 on the negative feedback side of IC7. I put circles around R52 and R59. These values only give IC7 (a TL061 operational amplifier an apparent gain of about 1.18 (Av = 1.18).

With a gain of 1.18 (almost unity, no gain at all) and with an input sensitivity of 1 VRMS, resistors R53 & R54 feeding the step-down series voltage regulator transistors T5 & T7 for IC7, will only see a maximum voltage swing of 1 Volt peak to peak to drive Mosfet Op Amp drivers A1 through A4 with.

This will then yield less than 10 Watts maximum by the amplifier into an 8 ohm speaker load, so either one or both of the values is not right.

I made a breadboard circuit utilizing the values of your schematic. I then tested 3 operational amplifiers, a TL061, a TL071 and a TL081 as they are all quite similar in specification from the manufacturer.

With any voltage at any frequency from 20 Hz to 20KHz applied to non-inverting pin 3 up to plus and minus 15 volts, quite nearly the same voltage appeared at the output of the operational amplifiers.

Seeing that operational amplifiers A1 through A4 are supplied the equivalent of plus and minus 10 volts, the output of IC7 should never exceed the supply voltage of A1 through A4 as although these specific devices are not prone to latch-up, they will be caused to operate very erratically, perhaps destroying the connected Mosfet or the loudspeaker.

Not being the designer of this amplifier I would have to say the gain set by R52 and R59 should be much more that unity but probably no greater than 10.

Therefore, what should these values be for the amplifier to operate correctly?

Next problem.

This involves the protection circuit, primarily Zener ZD5. This Zener is labeled and used as a 4.3 Volt Zener in conjunction with the ±18 VDC supply being OK (+V1 to –V1) applied to the base of Darlington T9, and is also used labeled as a 43 Volt Zener in series with ZD7, another 43 Volt Zener in conjunction with the ±50 VDC supply being OK (+V3 to –V3) applied to the base of T3. See your schematic below.

Since I cannot paste the image here, I will email it to you.

I put circles around ZD5.

In either case, if ZD5 is 43 Volts in the (+V1 to –V1) supply OK or if ZD5 is 4.3 Volts in the (+V3 to –V3) supply OK circuits, neither will not work.

So now that I have unequivocally detailed several problems with your schematic (since these problems were found, there may be other problems that I have not encountered as of yet), once again here are a few more questions.

Is there a problem with my circuit board? Since there are problems with the schematic, what was the circuit board created from?

What needs to be done to correct all this?

Since I unequivocally detailed several problems with the schematic, can you provide one that is correct?

I’ve shown you that the amplifier cannot function the way your schematic is drawn.

I guess I could spend a lot of time and breadboard the amplifier to substitute circuit components in and out and evaluate the performance of all the permutations. This might work, but it misses any symbiotic relationship intended by the designer between the parts. It would take quite some time and effort to properly do this.

Therefore, someone designed this amplifier. Only they know what they intended. Only they know of the changes that proved had to made for some reason or another. There must be better information somewhere.

I would like to build the amplifier correctly the first time. Not unsoldering components and trying others. To begin with it’s not good for the circuit board. Once power is applied, components can change value, tolerance or even become destroyed.

Once again. Please Help.

Sincerely,

Richard Borkowski
837 Lindley Road
Bronson, Michigan 49028
E-mail: Rborkowski@vbs.net

4

This kit has been available unchanged for over 15 years, before it was discontinued.
Numerous reviews have been written.
Supplied assembly instructions are correct.
Watch for last minutes additions on a separate leaflet, indicated as ‘NOTE’.
The diagram is indicational and subject to minor changes.
Sorry, due to the age of the kit, it no longer possible to get feedback from the designer.