Hi John,
The HCU04s have just turned up (I'd ordered N SOICs instead of D DIPs... D'oh! One of those weeks...), but first I tried an experiment...
I tried what I'd previously suggested... using the D500 to give the Isomagic S/PDIF lock to make it happy, and then feeding the Iso clock into the Discmagic as normal.
And hey presto - we have clock lock! One single relay click, the player is working normally, and the little LED lit up on the front.
So the Discmagic appears to be working perfectly... it's the Isomagic that has the problem. Precisely what is wrong with it, though, it is a tricky one...
- Tom.
The HCU04s have just turned up (I'd ordered N SOICs instead of D DIPs... D'oh! One of those weeks...), but first I tried an experiment...
I tried what I'd previously suggested... using the D500 to give the Isomagic S/PDIF lock to make it happy, and then feeding the Iso clock into the Discmagic as normal.
And hey presto - we have clock lock! One single relay click, the player is working normally, and the little LED lit up on the front.
So the Discmagic appears to be working perfectly... it's the Isomagic that has the problem. Precisely what is wrong with it, though, it is a tricky one...
- Tom.
Re: Re: Troubleshooting the Discmagic
I couldn't get used to the optical connection. It sounds odd to me, like I'm listening to someone else's CDP. I went back to the coax connector, but I installed a cheater plug on the DiscMagic power cable. This sounds better to me and it also stopped the dropouts, so I suspect there was a ground loop that has now been interrupted.
audiobomber said:
I couldn't get used to the optical connection. It sounds odd to me, like I'm listening to someone else's CDP. I went back to the coax connector, but I installed a cheater plug on the DiscMagic power cable. This sounds better to me and it also stopped the dropouts, so I suspect there was a ground loop that has now been interrupted.
Hi John,
I've spent some time tracing out the Iso board to try and figure out what is going on. I discovered that the oscillator is powered via. two outputs in parallel of an AC11000 NAND logic chip, and when there is no signal lock, the oscillator power is switched off.
Attached is the schematic I've managed to come up with through probing. No idea what the 42C70 is... small EEPROM?
The logic appears to be working as designed... or at least as the PCB has wired it! According to the truth table I just wrote out, only with a signal lock and the locked LED turned on (by whatever is responsible for that) will the oscillator be powered.
- Tom.
I've spent some time tracing out the Iso board to try and figure out what is going on. I discovered that the oscillator is powered via. two outputs in parallel of an AC11000 NAND logic chip, and when there is no signal lock, the oscillator power is switched off.
Attached is the schematic I've managed to come up with through probing. No idea what the 42C70 is... small EEPROM?
The logic appears to be working as designed... or at least as the PCB has wired it! According to the truth table I just wrote out, only with a signal lock and the locked LED turned on (by whatever is responsible for that) will the oscillator be powered.
- Tom.
Attachments
Tom,
Are you sure - thats not what I designed? - so unless Cambridge Audio changed the design for some reason - I cannot comment -
My design was Rev 1.0 - Copywrite 1996
the 42C70 is a PLL Multiplier x 1.5 I seem to recall (256fs from 8412 to 384FS for SAA7350 - thats why clock-lock is so much better - the 7350 operates directly from XO without the need for PLL.
Are you sure you have the correct pinouts for the 74AC11000? I cannot believe CA changed the design...
John
Are you sure - thats not what I designed? - so unless Cambridge Audio changed the design for some reason - I cannot comment -
My design was Rev 1.0 - Copywrite 1996
the 42C70 is a PLL Multiplier x 1.5 I seem to recall (256fs from 8412 to 384FS for SAA7350 - thats why clock-lock is so much better - the 7350 operates directly from XO without the need for PLL.
Are you sure you have the correct pinouts for the 74AC11000? I cannot believe CA changed the design...
John
John,
Yes... I've just re-checked the wiring with my DMM again, and they are connected as shown in the diagram.
It does look that way. Someone who never bothered to check the implications of the changes, I suspect... either that or there's a PCB error...
This is a PCB from an "S700" labelled Isomagic. It is marked "ISO REV 2.1", and the text at the top reads...
(C)1999. JOHN WESTLAKE
TO AL & LB
Straight from the TI PDF, downloaded a few days ago.
If you could suggest how it is supposed to be wired, I can make the necessary changes and hopefully have working clock lock...
Thanks,
Tom.
Yes... I've just re-checked the wiring with my DMM again, and they are connected as shown in the diagram.
It does look that way. Someone who never bothered to check the implications of the changes, I suspect... either that or there's a PCB error...
This is a PCB from an "S700" labelled Isomagic. It is marked "ISO REV 2.1", and the text at the top reads...
(C)1999. JOHN WESTLAKE
TO AL & LB
Straight from the TI PDF, downloaded a few days ago.
If you could suggest how it is supposed to be wired, I can make the necessary changes and hopefully have working clock lock...
Thanks,
Tom.
Tom,
I will be meeting with the CA guys in HK at the end of next week, so I will investigate more - I suspect that there must be another problem as the Clock Lock should still work even if the clock circuit requires signal Lock to enable its operation. Many of these units have been sold - so this must be an isolated issue.
There must be an issue with the DiscMagic - but what, I cannot say without seeing the unit. I can send you a replacement DOTX board (you will have to add U3 and its heatsink). The board's in Europe - so I cannot send until my return at the end of Oct.
If this does not solve your problem, then I recommend purchasing one of my new transports - these will be budget priced but will still out perform all transport on the market in terms of Phase noise performance and disc reading ability.
After my experience with the servo boards used in the CD4 (non SE), CD6 and DiscMagic I spent a great deal of time improving read back performance of CDR and CDRW of my new transport with a 2 stage attack - first a fully variable gain RF summing amplifier from the Laser Pickup and should the VGA reach maxim Gain, then the laser Power is automatically increased – the transport reads any disc you load into it – provided its not badly physically damaged.
John
I will be meeting with the CA guys in HK at the end of next week, so I will investigate more - I suspect that there must be another problem as the Clock Lock should still work even if the clock circuit requires signal Lock to enable its operation. Many of these units have been sold - so this must be an isolated issue.
There must be an issue with the DiscMagic - but what, I cannot say without seeing the unit. I can send you a replacement DOTX board (you will have to add U3 and its heatsink). The board's in Europe - so I cannot send until my return at the end of Oct.
If this does not solve your problem, then I recommend purchasing one of my new transports - these will be budget priced but will still out perform all transport on the market in terms of Phase noise performance and disc reading ability.
After my experience with the servo boards used in the CD4 (non SE), CD6 and DiscMagic I spent a great deal of time improving read back performance of CDR and CDRW of my new transport with a 2 stage attack - first a fully variable gain RF summing amplifier from the Laser Pickup and should the VGA reach maxim Gain, then the laser Power is automatically increased – the transport reads any disc you load into it – provided its not badly physically damaged.
John
Hi John,
Thanks for the offer of the replacement Discmagic board - it may help with some of the more recent issues I've been having (and that I hope Jonathan won't encounter!). The new transport sounds interesting, but I already have a very nice one for myself... uses a DVD-ROM mechanism and asynchronous reading... and double precision floating point DSP for oversampling...
What I've done to the Isomagic is disconnected 1A,1B,2A,2B from 3Y, and connected them to ground. I've then lifted 1Y and 2Y, and connected the two 2R2 resistors in parallel directly to the switch output.
This now means the clock comes on and stays on when the switch is flicked downwards, regardless of whether there is S/PDIF lock, and stops the vicious cycle of intermittent locking.
I now have working clock locking, it appears. I'm measuring pretty stable clocks at R26 in the Discmagic - slightly different frequency when clock locking is enabled...
16.934045MHz for clock locked
16.933255MHz for no clock lock
So it *seems* to be working.
Unfortunately, somewhere along the line, I appear to have done something bad to the Discmagic... there's a kind of distortion over the top of the audio regardless of clock mode unless I touch PCB ground with a finger, when it miraculously disappears... could this be due to the removal of the copper shield, and the ceramic caps on the servo board that were soldered to it?
If I feed from the D500 over the same coax lead, there is no such distortion. Gaaahh!
I'll just have to hope it goes away... really need to get on with the serious Isomagic modification now, otherwise I'll never get started before big life things take priority. I've attached a small picture to show how this Rev.2.1 used to look around the clocking area (before I altered it!).
Cheers,
Tom.
Thanks for the offer of the replacement Discmagic board - it may help with some of the more recent issues I've been having (and that I hope Jonathan won't encounter!). The new transport sounds interesting, but I already have a very nice one for myself... uses a DVD-ROM mechanism and asynchronous reading... and double precision floating point DSP for oversampling...
What I've done to the Isomagic is disconnected 1A,1B,2A,2B from 3Y, and connected them to ground. I've then lifted 1Y and 2Y, and connected the two 2R2 resistors in parallel directly to the switch output.
This now means the clock comes on and stays on when the switch is flicked downwards, regardless of whether there is S/PDIF lock, and stops the vicious cycle of intermittent locking.
I now have working clock locking, it appears. I'm measuring pretty stable clocks at R26 in the Discmagic - slightly different frequency when clock locking is enabled...
16.934045MHz for clock locked
16.933255MHz for no clock lock
So it *seems* to be working.
Unfortunately, somewhere along the line, I appear to have done something bad to the Discmagic... there's a kind of distortion over the top of the audio regardless of clock mode unless I touch PCB ground with a finger, when it miraculously disappears... could this be due to the removal of the copper shield, and the ceramic caps on the servo board that were soldered to it?
If I feed from the D500 over the same coax lead, there is no such distortion. Gaaahh!
I'll just have to hope it goes away... really need to get on with the serious Isomagic modification now, otherwise I'll never get started before big life things take priority. I've attached a small picture to show how this Rev.2.1 used to look around the clocking area (before I altered it!).
Cheers,
Tom.
Attachments
The noise went away when I put it back together again. Phew.
I've started on the major Iso modifications. After I've got sound (hopefully sooner rather than later!), I then need to figure out how I'm going to take advantage of local 384fs clock, when the PMD100 is running entirely off the CS8412 at 256fs, and the new converters don't use master clock!
- Tom.
I've started on the major Iso modifications. After I've got sound (hopefully sooner rather than later!), I then need to figure out how I'm going to take advantage of local 384fs clock, when the PMD100 is running entirely off the CS8412 at 256fs, and the new converters don't use master clock!
- Tom.
Interesting stuff.
i have the use of a 1999 Westlake Isomagic a the moment.
I run in from my PC via USB and a cheap (£8) USB to SPDIF converter I bought from CPC.
I've found the optical sounds much better than any electrical, with these two C.A. DACs anyway. Using a £70 Edirol USB to SPDIF converter as it had electrical and optical out, we found the less copper, or silver, between the converters electrical o/p and the DAC i/p the better. With just a BNC between them that equalled the optical into the (much better sounding and throw away priced (£30)!!!!!!) DacMagic 2, but optical was still a fair bit better on the IsoMagic.
Heres an odd thing. When I shut the 'pute down, the IsoMagic loudly 'howls and hisses'. Actualy all have to do to cause the alarming noise is unplug the optical i/p, the 'lock' LED goes out and I run for the source selector!
I've been using a 7350 for many years in a custon arrangement and it has a similar sound to the Isomagic, the mid and vocal is a bit too laid back and uninvolving. But some instruments are very natural.
The older (??) and cheaper un-moded DacMagic 2 by comparrison has treble, bass, interesting mid and stereo. In fact, when driven by a 'pute, it's rather good, all be it a tad harsh. Even on a no compramise system. But it sound carp on a poor transport. I came to realise how bad transports are! A variety of DACS we tried from the 'pute sounded not all that different really, but transports, a world apart.
i have the use of a 1999 Westlake Isomagic a the moment.
I run in from my PC via USB and a cheap (£8) USB to SPDIF converter I bought from CPC.
I've found the optical sounds much better than any electrical, with these two C.A. DACs anyway. Using a £70 Edirol USB to SPDIF converter as it had electrical and optical out, we found the less copper, or silver, between the converters electrical o/p and the DAC i/p the better. With just a BNC between them that equalled the optical into the (much better sounding and throw away priced (£30)!!!!!!) DacMagic 2, but optical was still a fair bit better on the IsoMagic.
Heres an odd thing. When I shut the 'pute down, the IsoMagic loudly 'howls and hisses'. Actualy all have to do to cause the alarming noise is unplug the optical i/p, the 'lock' LED goes out and I run for the source selector!
I've been using a 7350 for many years in a custon arrangement and it has a similar sound to the Isomagic, the mid and vocal is a bit too laid back and uninvolving. But some instruments are very natural.
The older (??) and cheaper un-moded DacMagic 2 by comparrison has treble, bass, interesting mid and stereo. In fact, when driven by a 'pute, it's rather good, all be it a tad harsh. Even on a no compramise system. But it sound carp on a poor transport. I came to realise how bad transports are! A variety of DACS we tried from the 'pute sounded not all that different really, but transports, a world apart.
It is March 2007, and I have been lucky to pick up a pristine set of Isomagic/Discmagic from a seller in Canada, and today I have played the set for the first time at my home in Tokyo. Toslink for Clock Lock and a high-quality BNC cable for the main link. I had a very nice Omar Sosa HDCD for the testing, but also played many other nice CDs (mostly jazz).
This combo sounds every bit as good (and perhaps even better) than its reputation. It really sings. I used a fairly simple system with a Yamaha AV receiver, but brand new (excellent sounding) Sonics Anima speakers (small monitor) by Joachim Gerhard (Germany).
Fantastic 3-dimentional sound with rich musical textures. Makes you forget the hi-fi and focus on the music. Moves you to tears if the musical content is worth it.
I have a lot of other worthy digital equipment around, and my heart is in analog playback with the best turntable, tonearm, MC cartridge combinations etc. Still this John Westlake masterpiece of a digital combination is remarkable, especially considering its price tag.
I will be checking the various hints for modifications and improvement for these units (especially the Isodac), however, my initial feeling is that what sounds so good stock, may actually be left alone except for minor tune-ups.
Cheers!
This combo sounds every bit as good (and perhaps even better) than its reputation. It really sings. I used a fairly simple system with a Yamaha AV receiver, but brand new (excellent sounding) Sonics Anima speakers (small monitor) by Joachim Gerhard (Germany).
Fantastic 3-dimentional sound with rich musical textures. Makes you forget the hi-fi and focus on the music. Moves you to tears if the musical content is worth it.
I have a lot of other worthy digital equipment around, and my heart is in analog playback with the best turntable, tonearm, MC cartridge combinations etc. Still this John Westlake masterpiece of a digital combination is remarkable, especially considering its price tag.
I will be checking the various hints for modifications and improvement for these units (especially the Isodac), however, my initial feeling is that what sounds so good stock, may actually be left alone except for minor tune-ups.
Cheers!
elecon said:
Congratulations on your find. I've owned a DiscMagic/IsoMagic for almost ten years, recently upgraded with an improved power supply. Fabulous sound and amazing value.
Anyway, I have a tip for you. I was using similar connectors to what you have, a Toslink (Kimber) for clock and a BNC (Ben Duncan) for data. First, I upgraded the plastic Toslink for glass optical, and it was noticeably better. Then, on John Westlake's advice, I changed the BNC coax for glass Toslink. At first I didn't get it, I thought the BNC was better, but on closer listening, I realized that what seemed to be extra detail with the copper wire was actually distortion. The sound was zingier, but without as much real detail. Here's an explanation from the master:
http://www.diyaudio.com/forums/showthread.php?postid=803344#post803344
I was pleased to see your recent post. I have just today paired the IsoMagic S700 with a DiscMagic 1, to take advantage of the clock-lock. I am just beginning to evaluate, but have an issue I feel is RF leakage / interference.
I see an early post stating "If you use Clock-Lock, then I recommend to use optical for both Clock (no Choice) & SPDIF - this helps Isolate RF between the units." I have done so using a .5m Kimber OFT-1 for clock and a 2m (too long) Thundercable for the digital link, and then vice-versa checking direction of cables. I until today used the IsoMagic with a Philips CD960 as transport, thru a Kimber OFT-1, with very good results so do not feel the issue is the IsoMagic nor the DiscMagic, particulary since the Discmagic works well with the clock-lock switched out. I then tried the supplied RG595/U coaxial for digital and the aforementioned clock connections, and find no such RF issue whether clock-lock is in or out. While the consensus seems to be that optical (or perhaps higher quality glass) is optimal for sound, there remain concerns re RF in other posts.
Since I seem to have an RF issue, has anyone been able to resolve that specific matter? If the issue is proximity of the units, has anyone tried various patterns of shielding? Does this issue continue if the components are separated by one shelf, thus losing IsoPlat isolation value? Are there any known issues with the clock-lock switch gear etc.?
Any info much appreciated.
Regards,
Barry
eleakist@allstream.net
p.s. I am near Ottawa, Canada on a clean, no industry grid, further improved by a dedicated circuit, highquality junctions and AC plugs, then a top of line PanaMax AC box (i.e. no known grid or ground issues and no similar issues in the past)
I see an early post stating "If you use Clock-Lock, then I recommend to use optical for both Clock (no Choice) & SPDIF - this helps Isolate RF between the units." I have done so using a .5m Kimber OFT-1 for clock and a 2m (too long) Thundercable for the digital link, and then vice-versa checking direction of cables. I until today used the IsoMagic with a Philips CD960 as transport, thru a Kimber OFT-1, with very good results so do not feel the issue is the IsoMagic nor the DiscMagic, particulary since the Discmagic works well with the clock-lock switched out. I then tried the supplied RG595/U coaxial for digital and the aforementioned clock connections, and find no such RF issue whether clock-lock is in or out. While the consensus seems to be that optical (or perhaps higher quality glass) is optimal for sound, there remain concerns re RF in other posts.
Since I seem to have an RF issue, has anyone been able to resolve that specific matter? If the issue is proximity of the units, has anyone tried various patterns of shielding? Does this issue continue if the components are separated by one shelf, thus losing IsoPlat isolation value? Are there any known issues with the clock-lock switch gear etc.?
Any info much appreciated.
Regards,
Barry
eleakist@allstream.net
p.s. I am near Ottawa, Canada on a clean, no industry grid, further improved by a dedicated circuit, highquality junctions and AC plugs, then a top of line PanaMax AC box (i.e. no known grid or ground issues and no similar issues in the past)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.