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 20th May 2020, 5:42 pm   #21
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: Wavetek model 146

I have finally got all the pitchn off.
The petrol soak was no good, it only worked when the pitch was very hot, to risky!!
I tried acetone, no good.
I found that by heating the assy up and flicking bits of pitch off worked. The pitch would not pick off when cold.
After several days I now have a fibre glass pcb with 2 10 pin diode array's and 14 transistors on it.
The arrays are TO-100 encapsulation, and have N 050 stamped on the top.
The foreward biased readings of the diodes is around 40 ohms + or - 3 or 4.
Copmparing the readings between the 2 array's I have found 1 o/c diode in one array.
I am now going to take out the array's from the pcb and draw the diode connections between the pins. As the 2 cans have the same number on them I am assuming that they are the same inside. When I have done this I can then draw the complete cct of the pcb.
The resistors all have 6 coloured bands on them. At this point I have not been able to relate the value to the colours, more work to be done here.
I have now got the 2 replacement tr's for the main gen pcb, so I will get to this soon.
If anyone can point me to 10 pin diode array's please do.
I checked the waveform into the block and it is a triangular wave, not a sawtooth as I stated earlier. Sorry about the confusion.
John.
theredhouseinn is offline  
Old 21st May 2020, 8:01 pm   #22
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: Wavetek model 146

Hi!

The nearest commercial equivalent I can find that matches your package, is the RCA CA3019, also has equivalents NTE905, ECG905, SK2346 and GEIC–290.

This device has four matched diodes in a bridge configuration, plus two isolated ones, and a substrate connection to pin 7.

See how you get on with these – these devices are still available and not outrageously expensive!

Chris Williams
Attached Files
File Type: pdf ca3019.pdf (87.7 KB, 93 views)
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!
Chris55000 is offline  
Old 29th May 2020, 2:09 pm   #23
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: Wavetek model 146

I replaced the 2 faulty transistors in the main unit, still no sine wave, wiper on a pre-set not making contact, moved it a bit, sine wave now at o/p skt, got to take out all pre-sets and clean them.
Aux unit.
I removed the 2 ic's from the little pcb and checked all the pins in order, both ways with the Avo on low ohms range. The result is the attached diag.. Although I have identified all the connections between pins as diodes, it is obvious that I have measured foreward biased junctions of transistors. I have also attached the cct of the little pcb. Apart from the 2 ic's the rest of the components are resistors, old style carbon composition types.
I have figured out the colour code of the very odd values of some of them.
As 8 of the resistors are 4 pairs of paralleled ones, so I will have to remove them from the pcb to check their value. I measured one pair and the resistance was varying a few kohms intermittently, so I will replace them with metal oxide ones.
Your help would be appreciated with ideas of the layout of transistors and diodes in the encapsulation.
Both of the ic's measured the same pin by pin, so, unless I have an intermittent I will assume that the 2 are ok.
I will put them back in the pcb along with the new resistors, solder it back in the aux pcb and pray. I will also be able to check voltages on the ic's.
John.
Attached Files
File Type: pdf Flu 2_000048.pdf (41.8 KB, 64 views)
File Type: pdf Flu 2_000049.pdf (88.9 KB, 67 views)

Last edited by theredhouseinn; 29th May 2020 at 2:11 pm. Reason: Forgot name
theredhouseinn is offline  
Old 29th May 2020, 7:19 pm   #24
buggies
Heptode
 
buggies's Avatar
 
Join Date: Jan 2009
Location: West Lothian, UK.
Posts: 760
Default Re: Wavetek model 146

Apart from the polarity of the diode shown between pins 1 and 9 in your sketch, I would say it ties in well with Chris5500 post 22 diagram of the CA3019.
There is a slightly different picture here https://www.mzentertainment.com/tech...rca_ca3019.gif
along with a note saying that pin 7 should be taken to a potential more negative than the peak signal applied to any other pin. This might account for all the additional diode junctions you have detected connected to pin 7.
So suggest re-check pin 1-9 junction.
__________________
George
buggies is offline  
Old 29th May 2020, 8:50 pm   #25
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: Wavetek model 146

Hi!

I'm in agreement with Member "buggies" comments regarding the polarity of the diode junction between 1 and 9 of the CA3019 · you've drawn it backwards in your first pdf!

I'll provide a redrawn diagram of your module in a neat format you can keep for your future reference, and I think you'll find 2 x CA3019s will repair your generator – it's possible that Wavetek could have selected them in matched pairs, but hopefully if you request two '3019s from the same batch–code number from the supplier when you order them, you'll probably get away with it – there's nothing to lose by trying!

Thank you for posting your work – I shall place a acknowledgement note on my redrawn diagram!

Chris Williams.

PS!

I'm puzzled about your reference to "14 transistors" – could you post good quality photos of each side of the PCB please?
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!

Last edited by Chris55000; 29th May 2020 at 9:18 pm.
Chris55000 is offline  
Old 30th May 2020, 10:47 am   #26
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: Wavetek model 146

John
Slightly puzzled following this discussion.
Either those devices are CA3019, or something entirely different.
I used that AVO method donkeys years ago when trying to identify faulty i/c's by comparing the suspect one with a known good one.
For starters, you appear to have 1 good and 1 bad, 2 good or 2 bad.
Let's assume 1 good and 1 bad.
Diodes usually fail s/c and transistors either open or s/c so any anomaly such as a 50% or more difference in a reading will give you a clue.
Identification of things just marked "50" might be tricky.
A catalogue from the period of manufacture might give you a clue to possible devices.
You might have op amps, dual transistors or something very odd.
Multiple diodes from one pin should indicate a power or ground connection.
So a good start is to identify Vcc and ground because both devices should share these two connections on the pcb.
Any other clues such as a logo on the can or is it just "50". Does the pcb have any markings?
Allan G3PIY
allan is offline  
Old 30th May 2020, 4:51 pm   #27
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: Wavetek model 146

Hi chaps.
First, the main osc is now working ok. All pre-sets have been cleaned and soldered back in.
I have achieved 0.18% dist. sine wave, the spec is 0.14 dist, that will do.
The aux section.
The pin detail on the diag. is copied from a similar 10 pin ic in a data book.
Pin 10 is by the little lug on the cap, then 1 to 9 are numbered clockwise round from the pin 10, viewed from the bottom of the ic.
Pin 7 is connected to the can, and no connection on the pcb.
There is no connection to pin 10 at all.
I have just realised that I left out a 49.9ohm resistor connecting the o/p to the chassis conection, modified cct attached.
vcc+ is connected to pin 5 of both ic's ans vcc- is connected to pin 8 of both ic's.
I have rechecked the diode connections between pin1 and pin 9 and the anode and cathode connections are correct.
All the devices have a forward resistance of about 40 ohms, indicating that they are good.
Allen. I have been using the Avo method of checking semiconductors since the 50's, and it never fails. I have an ESR meter for checking transistors, so I am going to sit down and see if I can make sense of this thing. The ESR meter will identify base, collector and emitter connections.
Chris. I will have a go at photoing the pcb, bit it is very small, about 1" X 1 1/2" and grubby due to the pitch encapsulation.
The logo on the top of the ic is "N 050", this is the only id on it.
I have had to order some small 15 kohm res for the pcb so reinstallation to the main board is pending.
John
Attached Files
File Type: pdf Flu 2_000050.pdf (93.8 KB, 47 views)
theredhouseinn is offline  
Old 30th May 2020, 7:05 pm   #28
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: Wavetek model 146

Something like an LM319 maybe John?
Allan
allan is offline  
Old 30th May 2020, 7:25 pm   #29
buggies
Heptode
 
buggies's Avatar
 
Join Date: Jan 2009
Location: West Lothian, UK.
Posts: 760
Default Re: Wavetek model 146

The circuit description in the instruction manual provided by Station X in post #2 ( para 3.2.4) states that the triangle to sine conversion is produced by a shaping network of complementary diodes pairs and resistors.
__________________
George
buggies is offline  
Old 31st May 2020, 9:51 am   #30
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: Wavetek model 146

I should have read the whole thread George.
I was distracted, finishing off my HP8640B overhaul (now complete) which was only possible after I got my download from Artekmedia. That original 146 document was really bad.
Interesting to hear of all those bad transistors in the 146. In my 98640B there were a couple in similar condition but due to corrosion where the legs met the plastic body.
On the other hand my Wavetek 2407 had bad transistors with low resistance shorts. Presumably after decades odd things happen to some transistors. Reminiscent of the OC170 series.
I also had a failure of a 74LS96 in the 8640B with internal shorts with no reason why it should have happened as its in the middle of other circuitry.
Presumably, after a long time either all the electrons get used up or all the holes get filled up?
Allan G3PIY
allan is offline  
Old 31st May 2020, 6:28 pm   #31
buggies
Heptode
 
buggies's Avatar
 
Join Date: Jan 2009
Location: West Lothian, UK.
Posts: 760
Default Re: Wavetek model 146

I have just re-read the thread again as was losing the place!
It is easier in USENET groups as the messages have proper indented threads.
I have a Wavetek 142 generator whose manual (from the web - but more legible) uses the same/similar 5 pin sealed block part number 142-011.
I rubbed out the substrate diodes from the reverse engineered sketch and the similarity is more obvious.
I also found that the model 130 used a 5 pin part no. 130-011 so probably all similar blocks.
Attached Files
File Type: pdf ic.pdf (23.8 KB, 52 views)
__________________
George
buggies is offline  
Old 6th Jun 2020, 2:15 pm   #32
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: Wavetek model 146

I have been working hard on this job.
I have drawn out the cct of the pcb, file attached.
I have given the diodes id no's, so that the ic diodes can be related to the cct diodes. File attached.
I went through the ic with the ESR meter and there are no transistors in the ic, just the 11 diodes.
I have also included photo's of 3 views of the pcb. attached.
I am waiting for the 316 ohm resistor to arrive before I can proceed any further.
If you spot any errors in the cct. please comment. I went round many circles to get this far.
So far the only faulty res I have found is one of the 179k's is o/c.
Out of the board the 4 15k's read ok, but when I checked them in cct, one was intermittent, so I may have had 2 faults originally.
John.
Attached Thumbnails
Click image for larger version

Name:	IMG_1105.jpg
Views:	83
Size:	74.9 KB
ID:	207804   Click image for larger version

Name:	IMG_1106.jpg
Views:	72
Size:	101.6 KB
ID:	207805   Click image for larger version

Name:	IMG_1107.jpg
Views:	61
Size:	66.0 KB
ID:	207806  
Attached Files
File Type: pdf Flu 2_000051.pdf (123.3 KB, 67 views)
File Type: pdf Flu 2_000052.pdf (45.1 KB, 45 views)
theredhouseinn is offline  
Old 11th Jun 2020, 11:09 am   #33
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Smile Re: Wavetek model 146

Success!!
All the resistors on the little pcb changed, the 2 diode array's refitted, and the assy. refitted to the aux pcb..
Switched on and sinewave on the scope.
Adjusted the 3 pre-sets for distortion and obtained 1.4%, this drifted to 1.5% after about an hour, after 2 hours this drifted back to 1.4%. Not bad for it's age.
So, in conclusion, with a lot of hours and great care, it is possible to repair the triangle to sine converter.
John.
theredhouseinn is offline  
Old 12th Jun 2020, 8:39 am   #34
allan
Pentode
 
allan's Avatar
 
Join Date: Oct 2005
Location: South coast near Ringwood/Christchurch, Dorset, UK.
Posts: 230
Default Re: Wavetek model 146

Well done John.
What's your next challenge?
Allan G3PIY

PS do the presets change the level of harmonics so is that a method for correcting distortion?

Last edited by allan; 12th Jun 2020 at 8:41 am. Reason: afterthought
allan is offline  
Old 13th Jun 2020, 1:12 am   #35
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: Wavetek model 146

I recently restored a Wavetek 154, which also uses a resistor diode array (A1 module) for sinewave generation.

Pots R85, R86, R88 interact a lot, so it took a spectrum analyser and many iterations to identify how to suppress the various harmonics and get sinewave THD down to 0.1 to 0.12%, but well worth I reckon as the default spec is 0.5% THD for that model.

Given the 14 resistors in the module, it is likely a similar diode-resistor circuit.
Attached Thumbnails
Click image for larger version

Name:	Sine shaper assembly.JPG
Views:	79
Size:	21.7 KB
ID:	208370   Click image for larger version

Name:	Sine generation schematicJPG.jpg
Views:	67
Size:	76.6 KB
ID:	208371  
trobbins is offline  
Old 13th Jun 2020, 11:43 am   #36
theredhouseinn
Hexode
 
theredhouseinn's Avatar
 
Join Date: Jan 2013
Location: Kelvedon, Essex, UK.
Posts: 272
Default Re: Wavetek model 146

The 3 pre-sets, R53, R55, and R28, are the 3 adjustments for sine wave distorsion.
R53 varies the current from the + line to the array, and R55 varies the current from the from - line to the array. R28 varies the signal level to the integrator ic prior to the sine converter. I don't have the luxury of a spectrum analiser, so I just wached the digital redout on the dist. meter, and adjusted the 3 pre-sets backwards and forwards to tune for min dist.
This week one of my Philips PM3240 scopes packed up. The mains indicator led is flashing and the trace is flashing on and off.
Looks like a ps fault.
The next project.
John.
theredhouseinn is offline  
Old 13th Jun 2020, 1:57 pm   #37
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: Wavetek model 146

Initially I did just one iteration of pot tuning to get lowest THD and thought well that looks like the best I can get.

But I then tried some selective harmonic reduction as well as not letting any one harmonic rule the THD. I found it quite tricky, but rewarding, to iteratively tune down particular harmonics in such a way that THD kept reducing - it was quite the merry-go-round.
trobbins is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:32 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.