F5 power amplifier

[Sangram]

There's no center tap on the primary Anil already saw that and his confusion originated from there.

I guess he'll have to use one on the live leg like I thought you would.

 

[Renron]

Yeah, I thought I remembered reading that......too bad.

Ron

 

[anilva]

There's no center tap on the primary Anil already saw that and his confusion originated from there.

I guess he'll have to use one on the live leg like I thought you would.

That's right. It is just one 220v primary winding. I will use one thermistor in the live part of the mains in series.

Thanks everyone.

 

[Blues]

NTC in parallel is big no-no

it's same as having just one

initially - one with slightly lower resistance will conduct more than other one , it will heat instantly , dropping resistance under one ohm , and other one will stay cool as ice

Try using the Thrermistor like this!

Ron

From the pov of both your local distribution transformer and your power supply transformer the 2 thermistors on the upper diagram are in parallel

 

[bobodioulasso]

In one way, they can be considered "in parallel" But they are not really : each one sees one winding only.

 

[Electrolastic]

I wonder if I can use any thermistors rated 10 ohm 6 Amp instead of CL-60, which I have no more after building a couple of F5's ?

 

[bobodioulasso]

why not?

 

[anilva]

Regarding R11 & R12 resistors

Hi,

I am grappling with the availability of 0.47 ohms, 3W resistors to be used as R11 and R12 near the Power transistors. The bias adjustment is measured across these resistors for 0.6v

Every 0.47 ohms resistor I tried in the market measures 0.8 ohms on my Fluke DMM. My question is should I use two of these in parallel to get 0.47 ohms or is it okay to use 0.8 ohms resistors instead. What should the voltage measurement be if that is the case across the resistors instead of 0.6v? I remember v=ir but do not want to attempt equations with the Master's design.

Thanks in advance.

 

[nigelwright7557]

Hi,

Every 0.47 ohms resistor I tried in the market measures 0.8 ohms on my Fluke DMM.Thanks in advance.

I would guess it is your multimeter at fault.

Mine has trouble reading small ohms values.

Try shorting out the probes to make sure it is zero.

 

[Sangram]

The resistors are probably fine. To be sure, measure voltage across them and not their value. You can do this pretty easily with a variable voltage source and one reasonably large resistor, the voltage drop across the two resistors should be in proportion to their value.

My meter has a 0.2 default reading, so -.47 ohm resistors read 0.7 ohms or more. (no second decimal place). If you are still not sure, use 1 ohm resistors and parallel two of them, 1 ohm is much easier to find and it measures roughly correct on a 200ohm scale, 0.5 is close enough to 0.47.

 

[AndrewT]

The resistors are probably fine.
.......... If you are still not sure, use 1 ohm resistors and parallel two of them, 1 ohm is much easier to find and it measures roughly correct on a 200ohm scale, 0.5 is close enough to 0.47.

if you use two 1r0 in parallel to give you a net 0r5, then before hand you can connect the two 1r0 in series and measure 2r0 plus your DMM error.
If DMM error =0.2ohms and you measure 2.3ohms then the series resistors add up to 2.1ohms.
Divide that resultant by 4 and you get 0r525

If you were to pass 190mA through the series connected resistors and set your DMM to read 199.9mV, you should measure ~190mV across each resistor. If they measure a different voltage that tells you how well or how badly they are matched.

 

[labjr]

I use an ESR meter for low ohm readings. Works very well.

 

[Sangram]

@anilva - I think Sreekant (soundsgreat on the other forum) is also somewhere here, and he had an LCR meter that you can probably request him to let you use. He was measuring up his cables and had some lo-ohm readings along with inductance etc, so you might want to ping him.

 

[anilva]

The resistors are probably fine. To be sure, measure voltage across them and not their value. You can do this pretty easily with a variable voltage source and one reasonably large resistor, the voltage drop across the two resistors should be in proportion to their value.

My meter has a 0.2 default reading, so -.47 ohm resistors read 0.7 ohms or more. (no second decimal place). If you are still not sure, use 1 ohm resistors and parallel two of them, 1 ohm is much easier to find and it measures roughly correct on a 200ohm scale, 0.5 is close enough to 0.47.

Thanks. I suspect it is the measuring error as well. In fact I have a professional LCR tester and looks like I need to find it in my attic and dust it off for use.

Cheers.

 

[anilva]

if you use two 1r0 in parallel to give you a net 0r5, then before hand you can connect the two 1r0 in series and measure 2r0 plus your DMM error.
If DMM error =0.2ohms and you measure 2.3ohms then the series resistors add up to 2.1ohms.
Divide that resultant by 4 and you get 0r525

If you were to pass 190mA through the series connected resistors and set your DMM to read 199.9mV, you should measure ~190mV across each resistor. If they measure a different voltage that tells you how well or how badly they are matched.

Thanks for the advice. Will try this.

 

[anilva]

Maximum power supply voltage?

Hi,

Just hooked up my power supply with a light bulb, thermistor, fuse et al.

The power supply fired up properly, with the 100w bulb coming on momentarily and going off. The PS is not connected to F5 boards yet.

I measured the AC rails first. To my surprise, the rail voltage shows 21.5V as against 18V that I was expecting (supposed to be a 2 X 18v, 500VA toroid). Then I measured the DC output of the PS. The rail voltages are showing +28V and -28V. All this ofcourse on no-load condition.

Is this DC supply rail voltage too high as compared to the recommended 24-25VDC. I have spoken to the transformer vendor and he was saying the transformer will do 18V each rail, on full load condition and that it is normal to show ~20-21VAC under no load.

I would like to know whether it is OK to go ahead and connect the power supply to the F5 board with supply rails of +28/-28V or should I go back to the transformer vendor for a replacement?

I am using standard unregulated power supply design suggested for F5 by NP.

Thanks in advance.

Anil

 

[Sangram]

That sounds about right, actually, keep the light bulb connected and connect the F5. Remember the F5 cannot be tested with music playing with a 100W light bulb. You cannot set full operating bias with a bulb connected either.

Keep the trimpots to zero condition, check that the light bulb goes off after a second, so you know the circuit is safe and there is no miswiring. Then remove the bulb tester, and start the biasing procedure.

 

[AndrewT]

go ahead and connect the rectifiers and smoothing capacitors.
Re-measure the off load voltages.
Then connect up one channel of F5 with output bias resistors set to minimum resistance (=minimum bias current).
Re-measure the Quiescent voltages.
Start biasing the output in stages.
You will find that the PSU voltage will drop by 1 to 2V by the time you reach full output bias.

 

[anilva]

@Sangram - Thanks. Will connect the F5 board with the bulb first and then remove if things are alright, before setting the bias. I have already set the multi-turn presets to zero ohms by measuring. Will report on the progress.

@Andrew - Thanks. I have already connected up the bridges rectifiers and the smoothing capacitors to the transformer. The DC rails are measuring +28/-28V without load, and hence my question. I was not sure if was exceeding the safe operating voltage of the design.

 

[anilva]

F5 powered up!

Thanks to Mr. Nelson Pass and the forum members especially Sangram and Andrew for their timely help, my F5 build has been powered on.

Everything has gone as per the book. No problem. Bias voltage stabilizing after few adjustments and nearly 1 hour of burn at 0.580V and the DC offset between 3mv and -0.5mv.

Tested with input shorted and no load. Will start listening tests soon. Heatsinks are nicely warming up to a point, where I cannot hold my hand for more than 5 seconds on them.

So far so good. Took quite a bit of time in collecting the parts especially the heatsinks and the toroid. Will post pictures after I do my listening tests.

Thanks and cheers.
Anil