(AK) classA HA/Preamplifier.

[Deleted]

May 12, 2013 5:58:23 GMT FritzS said:

Are using single contact-pins (like as in good DIP sockets) are a good choice for those components, which are changeable? After testing, this could soldered too.

I buy'd MRS25 resistors - but which are recommend to use HOLCO or similar resistors like this www.hificollective.co.uk/components/prp_pr9372_metal_film_resistors.html?

For C1 100pF I have film capacitor, polypropylene (Mouser 505-FKP2100/100/2.5),
for C2 68pF I have MICA (Mouser 598-CD15ED680GO3F) in my stock - are this OK?

Hi Friedrich
You could use those singles for the output resistors . I would use a normal good quality DIP 8 socket for the offset corrector I.C. but make sure that it's pins are long enough to protrude right through the 3mm thick PCB. The same applies to all terminal blocks. I would use normal .5W MF resistors for all except the LS352 emitter resistors, where .1% tolerance types would be an advantage.
Alternatively, you could get a couple of packs of that value (150 ohms) and match pairs for that location with your DMM.. The exact value isn't crucial, but being closely matched is worthwhile.
Due to the thickness of this PCB it isn't a good idea to fit sockets or pins that will need to be replaced later. A solder flux pen is also good to use for some of the through layer holes.
Double check everything you do with the layout diagram before soldering them in. You can bend the leads under the PCB to hold them in place while you confirm that everything is correct.
I find it best to do both PCBs at the same time with resistors etc. Insert a resistor on one PCB, then insert the same value resistor at the same location on the other PCB for example.
At a later date you may wish to try higher quality resistors at a couple of locations such as the 4.7K and 10K in the feedback network. Don't forget the 2.2nF to 4.7nF across the 1M resistor of the offset corrector under the PCB. The capacitors you mentioned will be fine. An OPA134 MAY give a lower DC out than an AD744, as well as be a smidgin quieter.
Kind Regards
Alex
N.B.

 The 1N5819 diode should be replaced by an SF11 or SF12 (or ES1D Surface mount type) low forward voltage diode, and the 82 ohm emitter resistor in the V.A.S replaced by 120 ohms. This gives improved front end tracking with temperature.

[FritzS]

Don't forget the 2.2nF to 4.7nF across the 1M resistor of the offset corrector under the PCB.

Whats the best value?
And what correct this cap?

An OPA134 MAY give a lower DC out than an AD744, as well as be a smidgin quieter

I will see wich I have in stock.

[Deleted]

Friedrich
The capacitor reduces wideband noise of the Offset Corrector opamp itself, due to the high gain used. The value isn't critical.
It's a bit of "icing on the cake."
Alex

[jonclancy]

I've installed said cap and the 100R op resistor. Also some pcb pins to make measurement/balancing easier.

So far, so good.

Into the system next week.

TTFN.

[Deleted]

May 25, 2013 18:51:29 GMT jonclancy said:

I've installed said cap and the 100R op resistor. Also some pcb pins to make measurement/balancing easier.

So far, so good.

Into the system next week.

TTFN.

Hi Jon
Make sure that the output leads from the terminal block to the output RCA sockets when used as a preamp, are similar to what you use in your RCA leads.
I decided to remove my little delay PCB at the rear as the Loudspeaker Protector already has a delay , and was surprised to hear the difference on DTV audio after replacing the thin screened cable leads that I had been using. Newsreaders voices etc. just sound more natural. ( I am still noticing the difference days later) I presume that it has something to do with the lower resistance on the earth side of the whole chain through to the amplifier.
The same applies to the earths from the JLHs to the individual channel PCBs. If your layout is such that the PSU lead length is considerably longer for one channel, try doubling up on it's earth wire.
That was an earlier surprise finding. As I was already using 7.5A rated wire, I didn't expect that.

Kind Regards
Alex

P.S.
For those still to make this work as a preamplifier, I would recommend using a rotary selector switch such as this from ebay.
www.ebay.com.au/itm/4P-5T-4-pole-5-position-Ceramic-Rotary-Switch-RF-Power-/120693443092?pt=AU_B_I_Electrical_Test_Equipment&hash=item1c19e3ca14
This will permit switching of the earth side of the inputs as well, which can wake a worthwhile improvement where the audio outputs of a couple of SMPS powered devices are plugged in.
It's fiddly to terminate the incoming cables, but gets over the need to use relay controlled switching, with their power supply requirements, and unless 2 separate relay PCBs are used, you will not be able to switch the earth side, as well as having degraded channel separation if you use a single 4 input relay switching type PCB from ebay..

[jonclancy]

Thanks for the tips, Alex! I'm currently running single source, so it's a relatively simple set up in my chassis.

My 2nd boards are settling in nicely and are set to within a few mV of where they should be. You have to be quick setting the balancing. It would be great to install a couple of sets of test sockets on the chassis and a hole for a thin screwdriver for tweaking. I suspect that my Rapid multimeter isn't perfectly accurate, mind you - fine on the 2V scale, but seems to wander on the 200mV scale.

Cheers

Jon

[Deleted]

May 31, 2013 8:09:15 GMT jonclancy said:

Thanks for the tips, Alex! I'm currently running single source, so it's a relatively simple set up in my chassis.

My 2nd boards are settling in nicely and are set to within a few mV of where they should be. You have to be quick setting the balancing. It would be great to install a couple of sets of test sockets on the chassis and a hole for a thin screwdriver for tweaking. I suspect that my Rapid multimeter isn't perfectly accurate, mind you - fine on the 2V scale, but seems to wander on the 200mV scale.

Cheers

Jon

Hi Jon
I think that most DMMs wander on the 200mV scale until they are connected in circuit.

Kind Regards
Alex

[Deleted]

Gentlemen we have mail

[Deleted]

Jun 5, 2013 14:02:02 GMT @shaun said:

...and thence to...

[Deleted]

Andalucia

I remember when Spain won the European cup and all of the players went straight to their regional flags and not the national one.

It was just great to see

[Deleted]

I have now installed Vishay TX2575 Bulk Metal Foil .1% resistors in the 4.7K and 10K feedback networks of the preamp/HA.
Initial results are quite promising, but I will need to make a final judgement after the nasty chesty virus I presently have, clears.
These resistors are quite fragile, and care needs to be taken when bending their leads to fit the normal size resistor spacing.
The hardest part of the exercise was fully removing the old solder from the holes due to the 3mm thickness of the original PCB.
I used a very fine PCB drill bit (.6mm ?) in a chuck designed for fitting into an existing drill's jaws, and slowly turned the drill by hand.
www.jaycar.com.au/productView.asp?ID=TD2010
This should not be a problem with the forthcoming version of the PCB.
Don't get me wrong though, as the 3mm thick original PCB looks sensational in comparison with normal commercial PCBs.

Alex

[Deleted]

Jun 30, 2013 1:27:54 GMT @sandyk said:

The hardest part of the exercise was fully removing the old solder from the holes due to the 3mm thickness of the original PCB.
I used a very fine PCB drill bit (.6mm ?) in a chuck designed for fitting into an existing drill's jaws, and slowly turned the drill by hand.

Alex

That's pretty much what I did the last time I attacked those boards, I just held the pcb drill bit between thumb and fore finger to do the trick.

I haven't got around to the new resistors yet, must get my arse out of the sling...

[jonclancy]

Nice one, Alex!!

I've got mine laids out ready on the bench (among parts of my "skip-find-standard" Philips CD960 that is getting a full cosmetic rebuild). Hope to have them installed in the coming days.

[Deleted]

I now have around 10 hours on the Vishay Bulk Metal Foil resistors in the 4.7K and 10K feedback locations.
Initially, I thought there was a little extra HF detail, but didn't trust my hearing yesterday. This afternoon I was aware of added HF ambience that was a little excessive ,and would normally have me back tracking on my most recent modification.. After several hours I was far less aware of this, and the overall balance seemed more natural.
It t looks like I will have to wait longer to know for sure the end result. One thing for sure though, is that I have never had excessive HF ambience previously disappear by itself.
Another possibility is that the lingering chest virus may be clouding my judgement.

Alex

[Deleted]

Random thoughts.

Does anybody who has constructed the Class A HA/Preamp have a pair of headphones with a 4 wire cable fitted instead of the original 3 wire cable ?

The thought crossed my mind, that as the HA is of dual mono construction, it could be interesting to try the individual headphone capsules connected directly to the L and R Output terminal blocks.
If there was any improvement, a 4 or 5 pin socket could replace the headphone socket, and a matching plug fitted to the headphone cable.

Alex

[Deleted]

An interesting plan/experiment. Unfortunately I'm not one of the lucky 4 core peeps.

[Will]

Jul 1, 2013 11:09:40 GMT @sandyk said:

I now have around 10 hours on the Vishay Bulk Metal Foil resistors in the 4.7K and 10K feedback locations.
Initially, I thought there was a little extra HF detail, but didn't trust my hearing yesterday. This afternoon I was aware of added HF ambience that was a little excessive ,and would normally have me back tracking on my most recent modification.. After several hours I was far less aware of this, and the overall balance seemed more natural.
It t looks like I will have to wait longer to know for sure the end result. One thing for sure though, is that I have never had excessive HF ambience previously disappear by itself.
Another possibility is that the lingering chest virus may be clouding my judgement.

Alex

Hope you're feeling better Alex.

Any more thoughts on how bulk resistors sound?

[Will]

Jul 2, 2013 22:18:32 GMT @cjarchez said:

An interesting plan/experiment. Unfortunately I'm not one of the lucky 4 core peeps.

I could be wrong, but I think that the K701's have 4 core cable as standard? Might be worth a try, but who is brave enough to take cutters to their 701's?

Oh, and congrats on the Admin, Chris!

[Deleted]

Jul 10, 2013 20:38:15 GMT Will said:

Jul 1, 2013 11:09:40 GMT @sandyk said:

I now have around 10 hours on the Vishay Bulk Metal Foil resistors in the 4.7K and 10K feedback locations.
Initially, I thought there was a little extra HF detail, but didn't trust my hearing yesterday. This afternoon I was aware of added HF ambience that was a little excessive ,and would normally have me back tracking on my most recent modification.. After several hours I was far less aware of this, and the overall balance seemed more natural.
It t looks like I will have to wait longer to know for sure the end result. One thing for sure though, is that I have never had excessive HF ambience previously disappear by itself.
Another possibility is that the lingering chest virus may be clouding my judgement.

Alex

Hope you're feeling better Alex.

Any more thoughts on how bulk resistors sound?

Hi Will
I agree with Shaun that they do sound a little better, but there doesn't appear to be much in it.
It will come down to how resolving the rest of your system is, whether it's worth going there or not.
Little things like this, and perhaps the Cree Si C diodes that Shaun likes, as well as perhaps better quality main filter caps in the 15W Class A will all give the whole system a small lift.
I got the feeling that the change , when playing the comparison CD that I made, using the old Oppo 981 caused the differences between these tracks with identical check sums to be even easier to pick.
As it's still school holidays, I haven't had the house quiet enough for indepth listening.
I do intend trying some MUR120 fast diodes (found cheap on ebay) in the preamp PSUs when they arrive ,instead of the nice SF11s to see if there is a minor improvement.
Unlike normal diodes, they also specify the forward recovery time.
Kind Regards
Alex

[Deleted]

I think you're right Will;

The wiring scheme for the AKG K701 is:

orange = right +
white = right -
yellow = left +
black = left -

There are already four wires in the cable, so all you need to do is cut off the 1/4" phone plug and attach two XLR cables. XLR has three connections labeled "1", "2", "3". Only "2" and "3" will be used. The "2" is the positive and the "3" is the negative

[jonclancy]

Alex, the bulk foils are apparantly well-known for taking their time with burn-in. I was told this when given the bypass scheme for the TKD stepper I have.

[Deleted]

Jul 10, 2013 22:18:39 GMT @cjarchez said:

I think you're right Will;

The wiring scheme for the AKG K701 is:

orange = right +
white = right -
yellow = left +
black = left -

There are already four wires in the cable, so all you need to do is cut off the 1/4" phone plug and attach two XLR cables. XLR has three connections labeled "1", "2", "3". Only "2" and "3" will be used. The "2" is the positive and the "3" is the negative

Do a temporary test using the bare ends into the Output terminals, and if you don't hear an improvement, fit the headphone plug again .

Nothing ventured, nothing gained ?

[Deleted]

BTW, Mike has already proved that the faster UF400x diodes give a worthwhile improvement with the Musical Fidelity HAs.
Perhaps it could be worthwhile trying the MUR120 there too ?
Alex

www.ebay.com/itm/50-x-MUR120RLG-MUR120-120-Ultra-Fast-Rectifier-Diode-1A-200V-Free-Shipping-/180856039635?ssPageName=ADME:L:OU:AU:3160

[Will]

Looking forward to hearing how you find the MUR120, Alex.

OK - who is going to take the cutters to their K701??

[jonclancy]

Not me!!

But, and it's only taken four weeks, I have today re-worked my boards with the bulk-foil resistors.

Seeing as this pair of boards were bob-on for bias and balance before the mod, I fired them straight up. Worked A-OK, and I also noticed some extra HF as Alex described - not much, but it is there. However, I know these resistors require a bit of burn-in and I also know that re-trimming the amps does make a difference to the sound.

I'll give them a while to settle in, reset the bias (I'm running 160mA) and re-balance the front end.

I'm getting close to living with this as a finished project. I want to reclaim my workroom back for more building/engineering, and I think that will mean a massive clearance of past projects, classic hifi and future projects that I won't get round to building. I'd like to try some Class A power amplification next and I have pretty much all the bits in stock to build a pair of Pass F4/5 etc.

It's been a while, but I really enjoyed a bit of soldering this afternoon - even if I need to get stronger reading glasses!!!! AAARRGGHHH!