Arcam Alpha mods

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Simon Hi

Long time sir - hope all is well

I stripped out the output stage as you can see and now I have no sound.
The i'v op amps have the correct voltages but the DAC chip get's a bit hot for my liking. Walter has instructed me to test the voltages on the DAC chip asap but before I heat her up again and do the test can you see something missing from my board that I should have left in place ?

Walter cleaned his off almost completely as you know and you've gone straight off the TDA too so I've nothing to compare my mess with.

Absolutely any suggestions welcome ( it WAS sounding very nice too :confused: )

Andrew
 
Hi Andrew,

Sorry to hear there's a problem. I would say that when I changed those 2nd op-amps I got no sound at all, so something could be a bit fussy on there. At the time I thought the op-amps were oscillating due to not enough gain (OPA627, which need gain). I guess in your case you've just removed one component too many but I don't know which off-hand!

Simon
 
Obvious question - is the ground of the new RCA jacks connected to the player's ground solidly?

Looking at the datasheet and the photo I can't really tell what's gone wrong, sorry. It's been a while since I've messed with the player and I'm a bit rusty. I've been away to several places recently so not been here with my system listening and tinkering as normal. I hope to get back to it.
 
Last edited:
They are connected to earth by a piece of wire which you can't see in this pic. ( underneath the kind of T Bar I created ) The wire connects to the same place the original old rca jacks were connected to - it has always worked fine this way. Having looked at the board again I think I may have stripped so many components away that an effective earth point is no longer there anymore !! Is that possible Simon ?
It certainly looks like that .. I think you cracked it ..

So, where would I take my RCA grounds from now instead ?

A bit rusty ?
You step in after ' several places ' and spot that ?
I've been staring at it for two days wishing I'd left it alone !!

I have two resistors to put back around the op amps, a cap to do and then I'll find another earth point and fire it up

Thank you very much !
 
I don't know if the earth could be missing because of removing parts but I'd say it's possible as I seem to recall there are some link bars that connect the earth in places, though I would be surprised if that's all that connected it to the RCAs. Worth a look though or just a continuity test using your multimeter.

The only other thing I could think of was that somehow the muting or de-emphasis parts could be causing an issue by being removed but I can't currently think why. I think removing those should be fine.
 
I don't know if the earth could be missing because of removing parts but I'd say it's possible as I seem to recall there are some link bars that connect the earth in places, though I would be surprised if that's all that connected it to the RCAs. Worth a look though or just a continuity test using your multimeter.

The only other thing I could think of was that somehow the muting or de-emphasis parts could be causing an issue by being removed but I can't currently think why. I think removing those should be fine.

I removed all muting and de-emphasis. No problem and slight improvement in sound. All you need is correct power to the 3 pins and the 3 inputs and it will sing. As I advocate, first thing check power at each of pins 15, 26 and 28. Then check continuity from pin 5 to your rca ground. Worst case, bypass the opamp and pick signal right off the tda at pins 6 and 25. If you have correct power in, and no sound then you have a problem with your input signal wires. I thougth I had fried the tda several times after hitting the 5 volts with 9 or 10. It is quite resilliant. It almost always comes back to a broken ground on input or output, or a broken signal wire. Check the solder joints on all your dem mods as well. Its easy to break something and simple to reflow solder. If its on the output signal, one channel will work. If its decoupling caps, you'll get distortion on one channel. Trust me I have done it all. Now that its all working its worth it. Worst case we could point to point wire a tda and be done with this pcb all together. 80% of the parts are superfluous. Clean power, clean signal, decoupling on the pins, and the DEM circuit will feed a discrete output stage magestically. All the rest of these traces are just antennas for noise. You will find it, do not dispair :)
 
Clean power, clean signal, decoupling on the pins, and the DEM circuit will feed a discrete output stage magestically. All the rest of these traces are just antennas for noise. You will find it, do not dispair
Nice post Walter.

As for robustness I once accidentally ran a TDA1541S1 with +12v on the dacs +5pin..for some time until I noticed the reg was too cold. That dac still works and measures just fine!

Oh, post above about sockets - I would not do it for two reasons:
1) Sockets mess-up HF behaviour and decoupling by increased loop inductance through longer paths. It's critical to best performance (not least because the DIL28 layout is already huge and have the power/GND pins spaced far apart already...)

2) Both this dac and the filter disspate significant amounts of power (TDA1541 about 0.72w; SAA7220 about 1.0W). It noticably helps keep temps down to sink some of it directly via board traces/ground plane.
 
On the servo I added 78/7912 regulators (installed flat for CDM clearance) to power the motor drive op-amp. I installed a 7912 and LM317 (configured for +12V) to power the TCA0372 (Focus drive and Radial drive op-amp). I installed a dedicated 7805 for the TDA8808/8809. I also installed decoupling caps where there were unused spaces on the PCB. Everything in this are must be quite flat to avoid fouling the mounting parts or swing-arm of the mech. See picture attached for example.

Hello Simon,

I have a Rotel RCD956BX which uses the same transport as the Arcam Alpha and also uses same two TCA0372s and TDA8808s that you describe. They are also awkwardly situated on the board under the transport.

I see from the photo that you laid the +ve and -ve 12v regs to the TCA0372 flat on the board to help clearance. Did you notice whether these regs ran hot in operation?

The Rotel schematics show these chips are fed from unregulated +ve and -ve 10v. Is Arcam the same and did you notice any issues when you upped the voltage to 12v? (I guess you have used a higher voltage PSU to supply the regs.

Thanks
Joe
 
Last edited:
Hello Simon,

Thanks for reply. I am not sure now whether to go ahead if the results aren't there. I have regs on every other chip, including the SAA7220, but to install regs on the servo/drivers will mean dismantling the whole player so I can get at them. The Rotel was never designed for tweaking.

I note that those who tried this mod on the Marantz CD63/67 mods thread report significant improvements, which is interesting that the same servo chips and voltages are involved.

Anyway, this Arcam Alpha mod thread is interesting because it seems that the Rotel RCD965BX has many similarities to your Arcam. The major difference is that the Arcam has a TDA151 DAC while the Rotel has a SAA7323 bitstream DAC. But both players have the 7220 and 7310 chips.

Do you have a service manual for the Arcam or know where I can get one? I would like to see how difficult it might be to install a 1541 DAC and output stage.

I have an LM317 based regulator on the 7220 with the ground wire soldere directly to the chips ground pin. The heatsink is quite large but it still gets quite warm. I took the ground from the 'Kelvin' ground point of the reg direct to pin 12 on the chip. You can see the bit of green cable poking out under the heatsink in the photo.

Post #1987 of the Marantz website has a description and jpeg of it.

Joe
 

Attachments

  • Heatsink.jpg
    Heatsink.jpg
    325.3 KB · Views: 184
Joe,

Maybe it's worth trying for yourself to be sure but I was quite disappointed not to hear any obvious gains after getting enormous improvements with the same mod on a CD63.

The SAA7220 is super-greedy!

The service manual's a little large for attaching here, though I'm having a go at splitting it into five parts. If that fails I'll upload it somewhere.

Simon
 

Attachments

  • Arcam Alpha Service Manual.part1.zip
    878.9 KB · Views: 70
Last edited:
Joe,

Maybe it's worth trying for yourself to be sure but I was quite disappointed not to hear any obvious gains after getting enormous improvements with the same mod on a CD63.

If you take a close look at the manuals you'll see why this is so. The Arcam is fundamentally a well -engineered player with regulated , if high-impedance, rails for everything already, and lots of decoupling of supplies ahead of the regs where supplies are shared. In particular, the disc servo (reg'd) drives the disc motor through a push-pull output stage powered off separate supplies. Merely swapping regs won't / can't improve that!

In comparison with which - the Marantz CD 43/53/63 architecture is an ill-considered, utter electrical mess with a noisy , and noise-sensitive, DAC lost somewhere in the middle of it - as I have opined before...
 
Hello Simon,

I can download the five files but cannot decompress them. I have the latest version of Stuffit expander on my iMac and this is supposed to open zip files, but no go. I have tried another couple of programs that claim to be able to open zip files but no go there also. Any suggestions?


In comparison with which - the Marantz CD 43/53/63 architecture is an ill-considered, utter electrical mess with a noisy , and noise-sensitive, DAC lost somewhere in the middle of it - as I have opined before...

Hello Martin,

Thanks for that information. From the sound of it, the my Rotel is closer to the Marantz set up than the Arcam, so I will now have a go at adding the regs. There are certainly none there now. The Rotel service manual shows unregulated + and - 10V for the TC chips, but on the +ve rail I measured 12.43V with no load dropping to 12.26V with the motor spinning. The -ve rail is 12.86V no load dropping to 12.59V with motor spinning.

I think this means that I use the existing rails with low dropout regs set to 10V, or perhaps even fixed 9V regs.

Thanks again

Joe

PS (and off topic) I see that you live in Bath. I was in bath in August last year and loved it. Captain Arthur Phillip who was the first Governor of Australia (or New South Wales as it was then,) retired to Bath and died there on 31st August, 1814. He lived at 19 Bennett Street, Bath.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.