advertisement


Naim clones on Ebay

If turns out to be the case, then there's a fault in the circuit, or it's badly designed in the first place.

Just as were the first NCC200s until the penny dropped.

Please don't ask me to sort this out though - I'm not in the business of rectifying design flaws for folks who rip off the hard work of others.
 
If turns out to be the case, then there's a fault in the circuit, or it's badly designed in the first place.
It's a consequence of calculation and demonstrates that there is a problem with the circuit as realised which ought to be understood before tweaking.

If Neil gives us the voltages around TR1, TR2, TR3 and TR4 perhaps the culprit will jump out.

Paul
 
If Neil gives us the voltages around TR1, TR2, TR3 and TR4 perhaps the culprit will jump out...

Paul,

are these the measurements you mean?

TR1: b - c 24.43v, b - e 0.595v

TR2: b - c 20.15v, b - e 0.582v

TR3: b - c 22.81v, b - e 0.604v

TR4: b - c 24.01v, b - e 0.645v
 
On Les' schematic R9 is 24K, on the Chinese clone its 27K. I don't have 24K so dropped in a 22 to try - the offset drops again to -85mV.

So I'll get some decent 24K resistors as I notice that's one place Les used 0.1%
 
I must have been a bit slow, but I just noted on the Chinese boards that the equivalent of TR1 is actually the inner of the two BC550s - not the outer one = assume nothing - check everything!

So I may have the higher HFE device in the wrong position!
 
OK, so just checked the HFE of the R1/R2 - it made no difference to the offset.

BUT thanks to Paul and Sparts - changed R1 out to 1.2K and the offset dropped to 49mV. Could this be the answer? And is there any harm in pushing just a little further to get well under the 50 limit?
 
Neil,

There are three main contributors to the DC offset,

1. The bias currents flowing into through the 24k and 2k7 into tr1 and through the 27k feedback resistor into tr2. If these bias currents aren't matched, or the total resistance on each side isn't matched, then you get an offset.

2. The currents flowing through the 100R resistors. If these aren't matched then, similarly, you get an offset.

3. A mismatch in the VBE of TR1 and TR2 goes direct to the offset.

You can estimate the offset by working out the currents and then calculating the voltages starting at the 0v at the bottom of the 24k, through TR1, the 100R pair, TR2 and ending up at the output. Ideally this adds up to 0.

I made a crude spreadsheet to do this. Using Neil's voltages, (swapping TR1 and TR2), assuming TR1/2 have an hFE of 200, and replacing the 24k with 27k I get 99mV. Putting the 24k back in drops this to 89mV. R1 to 1k2 produces 39mV. So not an exact estimate, but plausible.

I think the odd value is the VBE of TR4. I looked up the ZTX753 data and I think you'd expect nearer 600mV, unless the current is far too high. In which case I'd expect smoke.

With 'as published' resistors, a TR4 vBE of 0.6, TR1 hFE at 250 and TR2 at 200 I calculate the offset at 0.052. Which is in the ballpark for the design as intended. So some transistor selecting should bring it in.

Adjusting R1 to bring the input pair into better balance is a 'mod', it should bring benefits other than to the offset by improving the linearity of the input pair. But this would start to make it 'not a clone', and maybe there's a reason it is as it is.

Beware that this is all theory and chin stroking.

Paul
 
OK, so just checked the HFE of the R1/R2 - it made no difference to the offset.

BUT thanks to Paul and Sparts - changed R1 out to 1.2K and the offset dropped to 49mV. Could this be the answer? And is there any harm in pushing just a little further to get well under the 50 limit?

Go for it. Stability will be fine.

Glad you are getting some success!
 
I haven't had a lot of time the last few days - but I finally got around to messing with some resistor values this morning. With 1k4 in the R1 position I now get -1mV!

So now to the final assembly and listening tests.
 
Excellent Neil, you deserve some rewards for you effort here so I hope they're good and you feel like taking them further (teddy front end?)

Enjoy (I hope).
 
The amps are now finished and --- dare I say it---- plugged in.

Only one at a time and on a crappy garage speaker and playing direct from my iPod with no preamp - but I'm pretty impressed. Now they are only running on +/-28v - so I didn't expect too much, but the dynamics seem pretty good at first blush.

Anyway, they'll find their way back to their rightful owner who's going to use them in an active setup.
 
Hi,
I recieved my kit on Saturday morning, 3 days from Hongkong complete with instructions and diagram.
I'm now listening to it as I write. Kit was very easy to build with everything labeled correctly. Bias set to 8mv across 0.22 output resistor as per instructions, doesn't seem to drift.
I just lashed together a power supply using a transformer from an old Nad amp which was faulty, this gives me 40v dc +/- after rectifier and the two 10000uf caps per rail.
Previous to building this I was using a Gainclone on the computer upstairs, the differance is light and day.
This thing rocks, scale of the music is huge.
I need to build a decent power supply, maybe dual mono construction.
Some pictures here.
IMG_3884.jpg

IMG_3888.jpg

IMG_3889.jpg

IMG_3890.jpg

IMG_3891.jpg

IMG_3892.jpg


Duncan
 
Duncan - glad you had instant success - that certainly wasn't true in my case. Did you check the DC offset?

Neil
 
Can someone please help me with my boards?

I have the 140H broads, and everything is in its place. I have connected everything up, but I get no reading when I try to measure BIAS. The pot doesn't make any difference.

I also have a DC offset of approx -130mV. Both board measure the same.

I tried connecting it to a couple of old speakers, and I does in fact work, and it plays music. The strange thing about the amp, is that if I try to measure the BIAS points when playing music through it, I get a reading, very low (between 0,6-1,9 when measuring VDC on the 200m dial of my multimeter), but it varies with the musical content. The more dynamics, the higher the reading. This can't be right, right?

But I want to find out what's wrong. I have similar readings to neiljadman on all 4 transistors (Q1-Q4).

What can be wrong, or what can I do to try and fix it?
 
Hi,
Yes I did check DC offset -80mv.
So I changed the 620ohm resistor below the bc550 transistor to a 430ohm and ended up with -1.4mv.
Stuffed some big caps in there too.
IMG_3897.jpg


IMG_3895.jpg


Multisim.jpg


Duncan
 


advertisement


Back
Top