UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 4th Jul 2020, 10:31 am   #21
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

I think that the circuit is very similar and they cover the same frequency ranges, but the key differences I have observed so far are:

Type 63:
CW/FM2 (external FM?)
AM1
AM2
FM1
XTAL check
Male RF connector
Mains F.M. deviation = ±150
Dial has external adjustment

Type 63A:
30% AM
FM ±22.5kc/s modulated with 1000kc/s
FM ±75.5kc/s modulated with 1000kc/s
FM mains (50hz modulated), Ext FM, CW
XTAL check
Female RF connector
Mains F.M. deviation > ±150
X-Sweep Phase/Output
No external adjustment for dial

I still have the covers off so was able to access the crystal check socket from the rear. The output was approximately sinusoidal at a frequency of 5.14MHz and an amplitude of 250mVpp. The insertion point is about 3/8in wide but the connector inside looks like typical 4mm banana size so a millimetre or two smaller. I wonder whether anyone has or has seen a "Probe Unit P.N.10670" ?
Attached Thumbnails
Click image for larger version

Name:	xtal-check-output.jpg
Views:	72
Size:	74.8 KB
ID:	210248   Click image for larger version

Name:	xtal-check-jack.jpg
Views:	67
Size:	64.7 KB
ID:	210252  
WaveyDipole is offline  
Old 4th Jul 2020, 11:12 am   #22
OldTechFan96
Octode
 
OldTechFan96's Avatar
 
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,177
Default Re: Advance type 63A

The FM2 socket on the 63 is where the sweep output comes from.

Do you have a dual channel oscilloscope? You could feed the the output of the crystal into channel 1 and feed the RF output into channel 2 so you can 'see' the beating procedure. I don't know how well this would work, just a random idea.

It might be worth posting in the 'Wanted' section to see if anybody has a service manual for the 63A, or the probe unit.
OldTechFan96 is online now  
Old 5th Jul 2020, 11:43 am   #23
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

I experimented with connecting a high impedance crystal earpiece to the PHONES (CAL OUTPUT/XTAL CHECK) connector and turning the Frequency knob. A high pitch whistle with varying pitch akin to the heterodynes on a radio are emitted when the pointer is near a multiple of 5MHz marker on the dial. There is a null point in the centre which presumaly indicates when exact synchronisation is reached with a multiple of the frequency of the crystal. Since the crystal itself was a little of the 5MHz mark, the null points didn't quite line up but they were pretty close on all bands.

Note to self - make sure that the modulation knob is set to the XTAL CHECK position!
Otherwise a VERY loud whistle is emitted!

I have replaced one cracked Hunts 0.02 capacitor at the X-Sweep connector. I intend to replace the one at the Ext FM input as neither of these are critical to the oscillator. The remainder will require careful consideration. I think the two at the crystal check output will be safe to replace, but I think any others will need to be done one at a time monitoring the effects, or perhaps are best left undisturbed for now.
WaveyDipole is offline  
Old 8th Jul 2020, 3:38 pm   #24
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

I have been looking at the two capacitors marked on the attached circuit diagram. The capacitors used in these positions are Hunts Mouldseals.

C10 is a single 0.005 but in the position of C11 there are actually two capacitors: a 0.05 and a 0.005 in parallel. I have a couple of questions:

Firstly, is the purpose of C11 filtering or bypass? Is the presence of two capacitors meant to deal with two different frequency bands, or was this a matter of factory selected correction? The 0.05 capacitor has a pretty wide 20% tolerance. Both the 0.05 and 0.005 seems to be original factory fitted Hunts black Mouldseals.

I have a suitable 0.047 to replace the 0.05, but my second question is whether ceramic capacitors of appropriate voltage would be suitable to replace both of the original 0.005 paper capacitors? I have a number of decent 0.0047 or 4700pF at 1kv, but might the micro-phonic properties of ceramic disc capacitors pose a problem here? I rather suspect that it shouln't be a problem for C10, but am not sure about the C11 pair.

There is also the matter of the adjacent C33 which is a 0.04. This seems a rather specific value, but it is also a Hunts paper capacitor, so 20% tolerance or perhaps 10% at best for these black ones? It is not stated in the parts list or on the capacitor body. Would this have been specifically factory chosen?
Attached Thumbnails
Click image for larger version

Name:	circuit-01.jpg
Views:	70
Size:	110.6 KB
ID:	210558  

Last edited by WaveyDipole; 8th Jul 2020 at 3:44 pm.
WaveyDipole is offline  
Old 8th Jul 2020, 4:23 pm   #25
ronbryan
Octode
 
Join Date: Jan 2003
Location: Guildford, Surrey, UK.
Posts: 1,958
Default Re: Advance type 63A

C11 tunes the frequency of the 1kHz modulation oscillator. The 0.005uF cap was probably added in parallel to the 0.05uF 20% to get the frequency closer to 1kHz.

Ron
ronbryan is offline  
Old 8th Jul 2020, 4:43 pm   #26
Electronpusher0
Nonode
 
Join Date: Jun 2016
Location: Bognor Regis, West Sussex, UK.
Posts: 2,288
Default Re: Advance type 63A

Wow, I never realised quite how much work would be involved in restoring it! You are doing a magnificent job, it clearly went to the right home.
Peter
Electronpusher0 is online now  
Old 8th Jul 2020, 6:55 pm   #27
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

Quote:
Originally Posted by ronbryan View Post
C11 tunes the frequency of the 1kHz modulation oscillator. The 0.005uF cap was probably added in parallel to the 0.05uF 20% to get the frequency closer to 1kHz.
Ron, thanks. In that case I dare say that the value is probably not that critical although I can well imagine that at the factory they would have wanted to get as close as possible to the 1000c/s value stated by the control on the front panel. On this unit the pitch of the modulation tone changes slightly when switching between AM, 20% FM and 75% FM in any case.

Quote:
Originally Posted by Electronpusher0 View Post
Wow, I never realised quite how much work would be involved in restoring it! You are doing a magnificent job, it clearly went to the right home.
Having taken out all those screws to get to the innards, I figured that I might as well do a thorough job on it. As we know, those Hunts Mouldseals are unreliable and the odd one was cracked so it seemed sensible to replace all of them while the unit is dismantled, although the ones inside the oscillator unit do need a little thinking about. That is why I am proceeding one feature at a time and observing the impact of each change. In any case, I am in no hurry.
WaveyDipole is offline  
Old 9th Jul 2020, 9:46 pm   #28
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

I have now replaced the 3 Hunts and one Metalmec paper capacitor in the audio oscillator section. I used a quality ceramic 4700pF to replace C10. I used a barely in spec ceramic for C33 which measured about 4100pF. The replacement of C11 was interesting because the combined nominal capacitance of the two paper capacitors used was 55nF, although at 20% tolerance (as marked on the Metamec capacitor) could have swung this 11nF either way. Prior to replacement I had measured the audio tone frequency at approximately 680Hz. Having installed a 0.047 1kV rated foil capacitor, I found the audio frequency to be co-incidentally about 680Hz. I then tried various values in parallel to find a companion that would correct this to 1kHz. It turned out that I had to add another 22nF for a measured a total of about 68nF to get a 980kHz tone on AM, 1003Hz on FM. A test using 2 x 0.047 took the audio frequency to about 1.2kHz. It seems that significantly more than 55nF would have been needed to get close to the nominal 1kHz mark and even with the original capacitors at 20% over nominal this would have fallen short. Could it be that that swapping out C33 has shifted this somehow? I tried adding a few pF and even into the nF range to C33, but this had no effect on the output frequency of the audio oscillator. Tweaking only the value of C11 had any effect on this. In any case, both C11 and C33 have been replaced and the audio oscillator now works at very close to the 1kHz mark. Fortunately I could detect no micro-phonic effects after the substitution.

There is still a problem though, because there is no modulation when the mode control is set to the FM 22.5kc/s position. Some probing with an oscilloscope shows that the audio oscillator stops oscillating when the switch is set to this position and replacing the capacitors in the oscillator circuit has not solved this problem. Checking the resistors around the modulation mode switch I found a couple of resistors that had drifted well out of tolerance (a 39k that was well over 57k and a 56k that measures well over 74k). Replacements have been ordered. All switch points on the outer (rear) wafer of this switch seem to be making good contact, but I have yet to test the inner wafer which is rather more difficult to access.

I would like to ask a question regarding L7/C13. On the parts list this is described as "Crystal oscillator anode coil". Is this used for tweaking the frequency of the crystal oscillator? I am familiar with a trimmer capacitor being used to trim the crystal oscillating frequency, but less so with an inductor + capacitor being used in this way?

Last edited by WaveyDipole; 9th Jul 2020 at 10:15 pm.
WaveyDipole is offline  
Old 14th Jul 2020, 3:20 pm   #29
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

After a little break and dealing with a bereavement yesterday, I was able to get back to the generator today. The 39k and 56k resistors have been replaced and the audio oscillator now works on the FM 22.5kc/s modulation mode and the FM signal is being modulated. Deviation looks about a third of what it is when in 75kc/s mode so this looks about right.

The remaining Hunts capacitors at the crystal output have also now been replaced. These were a little tricky to get at and required a little patience but all now done and the beat oscillator on the crystal check mode has been tested and is working.

That completes replacement of all of the Hunts capacitors. The signal generator now works on all modes. It could probably do with an alignment check although frequency output seems pretty close if not spot on dial markings according to the frequency counter built into the oscilloscope. A minor tweak only would perhaps be required on some bands. I had wondered about adding an output (perhaps using an OP amp) to allow a frequency counter to be attached.
WaveyDipole is offline  
Old 14th Jul 2020, 9:05 pm   #30
OldTechFan96
Octode
 
OldTechFan96's Avatar
 
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,177
Default Re: Advance type 63A

You are making good progress!

Are you ready to try aligning a set? Alignment instructions often refer to a detector probe. Usually made with a OA XX series germanium diode.
OldTechFan96 is online now  
Old 15th Jul 2020, 12:27 pm   #31
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

No, I am not quite yet ready to align a set. The generator first needs to be calibrated/adjusted to ensure that the output frequency agrees with the dial.

When I measure the output frequency of the generator using the Philips scope, the frequency does not quite agree with a tuned in Yupiteru scanner. Of course, frequency measurement using the oscilloscope is not the most accurate and the Yupiteru scanner is more likely to be correct, but neither seem to quite agree with the generator dial although both are pretty close. It is also possible that such variance (e.g. 10MHz reads approx 10.3MHz) is within the tolerance of such an instrument? Ideally it would be useful to determine whether the 5MHz crystal is spot on frequency. If it is, then the adjustment could be made using the crystal oscillator output using the XTAL CHECK mode, but even that does not quite agree with the dial at present. The crystal frequency measures 5.13MHz on the scope.

I am hoping to receive a frequency counter soon so it will be interesting to compare but my dilemma at present is accurately determining a known frequency value.

Added to that is the fact that the manual states the generator frequency output may shift somewhat when the covers are off, but there are no access points when the covers are on so I think I need to compare readings with covers both on and and off to see how much (if any) variance there is.
WaveyDipole is offline  
Old 17th Jul 2020, 7:25 pm   #32
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

Anyone know what size knobs fit this sig gen please? None of my knobs for 6.3mm shafts seems to fit. These seems to be fractionally bigger in diameter.

The frequency counter arrived and the crystal seems to measure spot on 5MHz give or take a kilocycle or two. The HP3314A I have was also spot on up to 20MHz so it seems I have a reliable means to calibrate against. The beat frequency null points when using the XTAL CHECK mode also co-incide spot on with 5MHz intervals so at least I now know that I can calibrate using the crystal or the frequency counter and they will agree. Measurements are now making sense.
WaveyDipole is offline  
Old 17th Jul 2020, 10:46 pm   #33
Bufo Bill
Heptode
 
Bufo Bill's Avatar
 
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
Default Re: Advance type 63A

I have enjoyed reading of your efforts so far, and I am sorry to hear of your loss.
Regards from Bill.
Bufo Bill is offline  
Old 18th Jul 2020, 8:29 am   #34
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

Thank you Bill.
WaveyDipole is offline  
Old 19th Aug 2020, 3:10 pm   #35
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

I haven't updated the post in a while because of various things going on but I would like to get the project completed. It seems that it should be possible to tweak the frequency output to align with the scale (at least in the centre) by adjusting the LC pairs (C1/L1 throu C4/L4) on each range. However, it seems that the discrepancy between output and the scale markings gets bigger as one progresses to the far ends of the scale and I am unsure whether there is a means to set the top/bottom end of the scale as one does when one aligns a radio. There also seems to be no adjustment on the highest frequency range. If I can't figure out how to make the adjustments I will just leave it alone, but I was wondering whether anyone could suggest how I go about making the adjustments?

The manual seems to go into detail about how to align an FM radio, but appears to give no advice about adjusting the generator so I am a little stuck.
WaveyDipole is offline  
Old 19th Aug 2020, 11:12 pm   #36
OldTechFan96
Octode
 
OldTechFan96's Avatar
 
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,177
Default Re: Advance type 63A

When adjusting the LC pairs, where do you have the dial pointer set?

Signal generators don't seem to come with alignment instructions. It would have been a back to the service depot type job I think.

The calibration instructions for the Heathkit IG-102 does shed a bit of light on the process. To paraphrase:

'Each band should be adjusted near the low end of the band.'

If L5 (highest frequencies?) is a thick coil, it probably is adjusted by physically stretching/compressing it.

I hope that this is helpfull.
Attached Thumbnails
Click image for larger version

Name:	Screenshot from 2020-08-19 22-59-55.png
Views:	40
Size:	150.3 KB
ID:	213917  
OldTechFan96 is online now  
Old 20th Aug 2020, 6:23 am   #37
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,637
Default Re: Advance type 63A

"I am considering the possibility of doing the same. Did you do a straight with a 100ohm 10 turn pot or did you choose a different value? " Sorry can't remember or fine my notes for that job but at a guess I'd say I used something other than 100r, 100r 10 turn pots being rare. Think I might have used 500r or more likely 1k, but in my case I wasn't worried about maintaining the same Z out.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is online now  
Old 20th Aug 2020, 6:59 pm   #38
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

L5 is a wide metal strip (see photo) and doesn't seem to have any adjuster so I guess bending/compressing it would be the only way to make any adjustment although since the oscillator is working I dare not at present!

Thank you for the info regarding tuning to the low end of the band. It seems the unit does have a scale adjustment screw in the centre of the dial. Not sure why I thought it didn't. Looking at the instructions it seems for very accurate readings, one switches to crystal calibration mode, tunes to the nearest zero beat for a multiple of 5MHz and aligns the dial with that point by means of the adjuster. Of course, that requires a high impedance headphone to be plugged in to the probe socket. I can understand why it is buried deep within as it is to prevent unnecessary RF leakage but it means that I would need to construct something suitable in order to use that feature.

The accuracy of the generator is supposed to be within ±1% so I am going to do some more tests to determine how far out it is on each scale and whether there is sufficient adjustment on the dial to compensate for any discrepancy. Using the crystal the accuracy is supposed to be ±0.02% give or take. The crystal is within a handful of counts on the frequency counter so that should be good enough to work with.
WaveyDipole is offline  
Old 12th Sep 2020, 4:52 pm   #39
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: Advance type 63A

I had a chance to carry out some measurements today. The first test was to connect a frequency counter and take readings at convenient top, mid and bottom points on the scale. Most were within the designed 1% tolerance, but there were a handful of anomalies:

- the 230-100 range reads high on over 1% across 2/3 of its scale. At 0.77% high the top end is rather less is out, but this seems to suggest that the whole range is slightly shifted.

- the high end of the 100-50 range reads under by 1.95% although the mid and bottom points were almost spot on so there is a slight problem with linearity.

- the 50MHz end of the 50-25 range read over 1% low, the mid point was spot on and the low (25MHz) point was 0.72% high suggesting that the range is slightly contracted.

The next test was to see whether the scale adjustment could compensate for the lack of alignment at the far ends of each scale using the method described in the manual to make higher precision measurements. The frequency of the crystal read 5000.11 on the counter which is equivalent to 0.0022% of the 5MHz nominal value and so well within the stated ±0.02% tolerance.

Adjusting the scale position to match the beat of the crystal oscillator on both ends of the 25-13 range (i.e. at 25MHz and 13MHz), the adjustment fell just short of the alignment point at either end, but fell significantly short at the bottom end of the 230-100 range. Precise adjustment as described in the manual could, however, be made for the scale end-points of all the remaining ranges.

Overall, it seems that the instrument seems to perform to within the 1% tolerance specification which is nice to see. Where it falls short, the error is relatively minor considering the age of the instrument and components within. It should also be noted that these measurements were done with the inner cover removed to facilitate access to the crystal beat frequency signal. The manual does state that there might be some minor change when the cover is removed.

If I had a suitable probe then I could test with all covers in place as would be the case during normal operation. In the event, I will probably have to cobble something together. I thought about fitting a mono 3.5mm jack on the outer casing but realised that this would mean bringing the beat oscillator signal out through two layers of carefully designed RF shielding. Since the instrument is designed to deliver signals at the μV level, any RF leakage at even the lowest order of magnitude is something to be avoided. I believe that the curious design of the probe with its signal pin having to pass through three layers of shielding before making contact with the signal output port was with this in mind. The probe would be inserted to set up the scale for precision and then withdrawn during actual use of the instrument. In that way there would be no chance of even the slightest inadvertent RF leakage via any connectors or associated wires while the instrument was in operation.
WaveyDipole is offline  
Old 14th Oct 2020, 1:29 pm   #40
OldTechFan96
Octode
 
OldTechFan96's Avatar
 
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,177
Default Re: Advance type 63A

I noticed that the service manual for the Advance SG63E includes instructions for a full RF alignment.

Pages 8 and 9. The instructions probably can be adapted for your Advance 63A, WD.
OldTechFan96 is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:38 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.