Dragon - left channel high end severe roll-off

[tcp100]

So today I went over to Perry's and we put my Dragon through the wringer.. Most things looked good - W&F is solid, NAAC seems to work quickly and without any problem, and everything mechanical looks quite solid.

What we did run into, though (and I'm hoping someone like Nakdoc can give some pointers) is that the left channel rolls off DRASTICALLY from about 8-10khz and up. Almost to the point where at 15khz we were seeing a 5dB difference between channels. The right channel seems to have pretty solid response all the way through - so it's isolated to just the left.

The level adjustments at 400hz worked fine on both channels - this only happens in the high end.

We tried adjusting the azimuth and the gain pots on the board, and could never get it close to dialed in - so we're thinking it's something in the PB amp section. A cap issue perhaps? It's always the left channel - regardless of playback direction - and the problem gets more pronounced as the frequency goes up. We even listened to a few songs on the Nak test tape, and you can definitely hear a lack of high end in the left channel, too.

One other small quirk was that my Dragon did not respond to the instructions in the service manual to hold down fwd/rev with no tape and the cassette door open. In the manual, it says this should show a slight movement of the NAAC. With mine, there was no movement, but it was adjusted dead center. Both of Perry's dragons moved, and you can hear the NAAC move a bit. NAAC worked fine though and landed right where you'd expect when playing.

(I'm of course quite more concerned about the playback level issue.)

The PB level issue would explain what Perry saw when I sent him the Phoenix tape with the right level pegging, and the problems I was having dialing in the bias and levels.

Any tips, or ideas on where to start looking?

[jazzgene]

So today I went over to Perry's and we put my Dragon through the wringer.. Most things looked good - W&F is solid, NAAC seems to work quickly and without any problem, and everything mechanical looks quite solid.

What we did run into, though (and I'm hoping someone like Nakdoc can give some pointers) is that the left channel rolls off DRASTICALLY from about 8-10khz and up. Almost to the point where at 15khz we were seeing a 5dB difference between channels. The right channel seems to have pretty solid response all the way through - so it's isolated to just the left.

The level adjustments at 400hz worked fine on both channels - this only happens in the high end.

We tried adjusting the azimuth and the gain pots on the board, and could never get it close to dialed in - so we're thinking it's something in the PB amp section. A cap issue perhaps? It's always the left channel - regardless of playback direction - and the problem gets more pronounced as the frequency goes up. We even listened to a few songs on the Nak test tape, and you can definitely hear a lack of high end in the left channel, too.

One other small quirk was that my Dragon did not respond to the instructions in the service manual to hold down fwd/rev with no tape and the cassette door open. In the manual, it says this should show a slight movement of the NAAC. With mine, there was no movement, but it was adjusted dead center. Both of Perry's dragons moved, and you can hear the NAAC move a bit. NAAC worked fine though and landed right where you'd expect when playing.

(I'm of course quite more concerned about the playback level issue.)

The PB level issue would explain what Perry saw when I sent him the Phoenix tape with the right level pegging, and the problems I was having dialing in the bias and levels.

Any tips, or ideas on where to start looking?

Not sure if this is related but my 680zx showed a hint of less treble in recordings. I saw that I could not dial in rec cal levels and I wanted to get the bias just right for the tape types I use.

Long story short, after a full electrical calibration with the service manual, I got everything spot on. The recordings are pristine now and highs are back.

So I think maybe your Dragon needs just this. Did you check to make sure the PB amp level is correct? On the 680ZX, a 200nWb/m 400hz is at 100mV.

[perry]

This was with playback using level tapes. Bias adjustment doesn't come into play. Once PB is good, then REC adjustments are done via bias adjust. We were thinking orange cap .....

[A.N.T.]

As this HF problem is the same on both forward and reverse playback, it is unlikely to be the head. Have a look at the output noise spectrum playing a blank tape - the difference between channels should be clearly visible. Double check if the spectrum is the same for either forward or reverse play. Check the Dolby chip for the right channel (IC203) and components around it for a start.

Alex

[tcp100]

As this HF problem is the same on both forward and reverse playback, it is unlikely to be the head. Have a look at the output noise spectrum playing a blank tape - the difference between channels should be clearly visible. Double check if the spectrum is the same for either forward or reverse play. Check the Dolby chip for the right channel (IC203) and components around it for a start.

Alex

Thanks for the info Alex!

Not to be ignorant, but would this still be the case if we had Dolby completely off throughout?

And why the right? The right seemed to have a pretty flat response all the way up to 18khz (we didn't go any further.) In other words, it looked "normal" to us. The left went down to almost nothing as we went up that far.

Not countering your advice, just curious and trying to learn. :)

Perry, I don't think we checked just the straight noise spectrum on a blank tape. I'll have to try that when we get back. Good point; that'll give us some indication as to where in the path it is.

If the problem is the dolby chip, is that an obtainable part?

[A.N.T.]

Thanks for the info Alex!

Not to be ignorant, but would this still be the case if we had Dolby completely off throughout?

And why the right? The right seemed to have a pretty flat response all the way up to 18khz (we didn't go any further.) In other words, it looked "normal" to us. The left went down to almost nothing as we went up that far.

Not countering your advice, just curious and trying to learn. :)

Perry, I don't think we checked just the straight noise spectrum on a blank tape. I'll have to try that when we get back. Good point; that'll give us some indication as to where in the path it is.

If the problem is the dolby chip, is that an obtainable part?

Sorry, I've misread the channels in your original description! *sigh*. Then it is IC103 for the left channel. My logic is simple - the outputs of the PB amplifiers are switched to the same Dolby chip for both forward and reverse play. As it is very unlikely that both left sections in the PB head are bad, it is the part after the CMOS switch that is possibly faulty and that leads to the Dolby circuit with all its numerous capacitors. Most likely it is not the chip itself but one of the surrounding caps.

Alex

[perry]

As Chris had already been over for a few hours, where we pretty much had already nailed his K-1020 down, (damn motor rpm variations not withstanding) we just didn't have the time to trace the signal paths, which would have been the next step. Just locate both channels paths at major points on the main logic board while using the reference tape at 15kHz until we found where the voltage dropped. As you all know, the Nak Dragon I/O board is packed, and not the easiest to follow through. I or Chris will trace the paths out on the PCB diagrams, using the schematic and note major test points along the way. What we were hoping for was someone that could say "Been there, done that, look at cap umpty squat". All signals go through the Dolby board whether switched in, or not, so that is a relatively easy place to check signals. Still good before the Dolby board, then it is most likely in the Dolby board that the problem lies. Bad before the Dolby board, and it is most likely the PB amp. We were getting a little tired and not thinking as clear, by the time we got this far. Good advice, thanks, Alex.

[Nakdoc]

So today I went over to Perry's and we put my Dragon through the wringer.. Most things looked good - W&F is solid, NAAC seems to work quickly and without any problem, and everything mechanical looks quite solid.

What we did run into, though (and I'm hoping someone like Nakdoc can give some pointers) is that the left channel rolls off DRASTICALLY from about 8-10khz and up. Almost to the point where at 15khz we were seeing a 5dB difference between channels. The right channel seems to have pretty solid response all the way through - so it's isolated to just the left.

The level adjustments at 400hz worked fine on both channels - this only happens in the high end.

We tried adjusting the azimuth and the gain pots on the board, and could never get it close to dialed in - so we're thinking it's something in the PB amp section. A cap issue perhaps? It's always the left channel - regardless of playback direction - and the problem gets more pronounced as the frequency goes up. We even listened to a few songs on the Nak test tape, and you can definitely hear a lack of high end in the left channel, too.

One other small quirk was that my Dragon did not respond to the instructions in the service manual to hold down fwd/rev with no tape and the cassette door open. In the manual, it says this should show a slight movement of the NAAC. With mine, there was no movement, but it was adjusted dead center. Both of Perry's dragons moved, and you can hear the NAAC move a bit. NAAC worked fine though and landed right where you'd expect when playing.

(I'm of course quite more concerned about the playback level issue.)

The PB level issue would explain what Perry saw when I sent him the Phoenix tape with the right level pegging, and the problems I was having dialing in the bias and levels.

Any tips, or ideas on where to start looking?
First,
do not turn any adjustments! The only "azimuth pot" is in the NAAC adjustment section. There is an PB azimuth adjustment on the head block. Let's agree to keep good control of what the names of things are.
The reason not to turn things is that the Dragon is so interactive as far as the NAAC controls are concerned that I'm not even going to pretend I can talk someone through the procedures (which by the way, have errors in the service manual).
HF loss, both directions in one channel only is almost certainly a tape path problem. Do you have access to a mirror cassette? Carefully remove the pad lifter and try the playback test again. If the problem is gone then you likely need a tilt adjustment. If it is there, get good light and a magnifier and clean that stubborn sneaky dirt off the head.

[perry]

I've noticed some of the errors (they actually tell you to record over a level tape in one place!). We KNOW this deck has been "fixed" by another tech(s), which is why we are woking on it. On the Dragon, if both left channel gaps are off due to contact. (and we checked with more than a few tapes) that would mean that both outside edges have min contact. (tape channels are laid out FL-FR-RR-RL out to in), so I guess it COULD be the pad lifter slightly out of whack lifting the both outside edges somehow. We did not think of this (gotta love this place), but easy enough to check if it is easy enough to remove the pad lifter. How to do this? It would also be easy to check the signal level right before the the head amp of each channel at frequency. We had the NAAC disconnected when we did this test (CN-5), as well, so NAAC was not a factor.

[tcp100]

First,
do not turn any adjustments! The only "azimuth pot" is in the NAAC adjustment section. There is an PB azimuth adjustment on the head block. Let's agree to keep good control of what the names of things are.
The reason not to turn things is that the Dragon is so interactive as far as the NAAC controls are concerned that I'm not even going to pretend I can talk someone through the procedures (which by the way, have errors in the service manual).
HF loss, both directions in one channel only is almost certainly a tape path problem. Do you have access to a mirror cassette? Carefully remove the pad lifter and try the playback test again. If the problem is gone then you likely need a tilt adjustment. If it is there, get good light and a magnifier and clean that stubborn sneaky dirt off the head.

I think the "pots" we were talking about were the 4 PB level ones in the playback section, (Labeled FWD RCH/LCH, REV RCH/LCH) rear right of the board with the front of the deck facing us. We tried evening out the levels there to no avail; those were the only "pots" we touched, and I believe brought them back to where they were. We left them so the levels were balanced at 400hz as described in the SM.

We did try aligning the PB azi per the service manual with CN5 disconnected, but that's the only other adjustment we attempted.

I know you said the SM has issues - but we didn't get too deep.

We tried cleaning - that helped a TINY bit, but nowhere near close to working. Hmm. Tilt would make sense, though. I do not have a mirror cassette, unfortunately.

Dare I try it with a cut-out clear cassette, or is this something best left for the experts?

[Nakdoc]

A cutout cassette is better than nothing. Look for ripples as the tape travels. You can make your own mirror tape if you have a 1/4" by 1 " mirror.
The pad lifter slides off easily. Do not use a magnetic screwdriver! The idea is to let the pressure pad maintain tape/head contact. If the left channel treble is ok, then you've proven that tilt/guide height/torque problems are causing tape travel problems.

Since cleaning helped, clean it some more! With the pad lifter off it will be much easier to clean!

[perry]

Well, we sure didn't know it slid off. It may just not be on right. Also, the outsides of the head (both left channels) are hardest to clean with the lifter in place, so that MAY be all it is! May also be the issue with the NAAC on my other Dragon. I'll look as soon as I get a chance. Thanks, Doc!

[tcp100]

Well, we sure didn't know it slid off. It may just not be on right. Also, the outsides of the head (both left channels) are hardest to clean with the lifter in place, so that MAY be all it is! May also be the issue with the NAAC on my other Dragon. I'll look as soon as I get a chance. Thanks, Doc!

I was thinking the same thing.. When I get home on Friday I'll pop the lifter off; here's to hoping I see two lines of crud on the outer edge of the head.. If not, I'll try watching the path with my cut-out cassette.

I'm dying to get my hands on a real mirror cassette - as I'm having tape path issues with my Sony decks as well.

[perry]

Well I'll be a Naks uncle...they just pull straight up and off, easy as pie. Makes head cleaning much easier. Definitely worth checking the NAAC without when I get a chance.

[tcp100]

Well I'll be a Nak's great grandaddy. Nakdoc, I just finally got around to doing a full FR test on my dragon with the pressure pad lifter removed - and I'll be damned if the thing isn't flat and equal on both channels with the lifter off!

Now I'm really feeling a need for that mirror cassette from A*Bex/medio-rite. Any hints on what to adjust now that it seems like it's a tape path issue with the pad lifter on? Could the lifter itself just have been askew?