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 28th Dec 2021, 11:35 pm   #1
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Gould 1504 repair

Last week, I could not resist to prevent two defective Gould 1504 scopes that were beyond economically viable repair from being thrown away. However, these scopes are filled with ICs, contrary to the IC-less analog 1980s stuff I usually work with. So, a lot of interesting things to learn during the repair.

I was not able to find any information about this scope online but it seems to be a version of the Gould 1604 with less memory and without any options. Both scopes had been sitting in a lab for decades unused, never having been recalibrated or opened, so the original cal stickers were still intact. After a quick check for shorts, I tried to switch the units on and did not get any magic smoke. One unit sometimes comes on partially, but has no screen, the other one comes on, but the screen content is shifted halfway to the bottom of the screen and the content has a strong circular blur on it, like some filtering caps are bad. The front panels are sometimes partially functional on the first unit and mostly functional on the second one. Judging from the amount of dust one of the units must have seen much more use than the other.

A quick checkup of the scopes revealed two blooming Varta MemPack batteries but no obvious other faults. Fortunately, due to the scopes being stored dry, unused and without being moved, none of the crystallized battery mucus made it onto the mainboards. So I removed the 4-channel acquisition boards, carefully extracted the batteries to prevent further damage and cleaned up the surroundings. Does anybody know, what the battery is used for in this scope? Cannot possibly be for storing the firmware, since the second scope would not boot up with a dead battery then.

After I found a very bad scan of the 1604 service manual online, I started probing the power supply outputs. Both supplies have several bad rails, with only the +12V rail(orange cable) being in spec and the +5V rail(red/brown cable) at 5.4V too high:

Code:
Color       Spec        Scope1       Scope2
=========================================
orange       12V        11.95V       11.9V
pink        -12V        -7.3V        10.8V
yellow       -5V         4.6V        -5.1V
white       180V       147V         165V
red/brown     5V         5.4V         5.4V
blue          FB         5.0V         5.0V
purple       75V        50V          76.8V
3+4           0V         0V           0V
grey          0V         0V           0V
The 180V, 75V, -12V and -5V rails are also out of spec, but not dead. I decided to leave the better unit with the working screen as is and to repair the other unit first.

With the power supply disconnected from the mainboard and applying a 20mA load on the rails, all rails appeared to be working fine without any significant ripple. So I first suspected something to draw too much current from the -12V or -5V line and the power supply collapsing as a result. However, I could not observe excessive voltage drops on the R70(BrightUp), R55 and R60(X-Amp) resistors which were suggested to check in the service manual and I could not reach R41(Y-Amp) with my DMM probes. Does anyone know the best way to extract the CRT board from this scope?

On a second thought I checked the six blue 10µF caps on the power supply and they turned out to be quite low and in need of a replacement. I also noticed that the MUR415 diode D2 had lost its markings and was covered in a black slimy substance, which I had never seen on a diode. Unfortunately, it turned out that both C10 and C23, despite not being off-valued, had leaked electrolyte and needed to be removed. Below the caps was only moderate damage to the soldering pads. Also, the electrolyte must have evaporated and attacked the label on the diode. Upon trying to check for electrolyte damage below the neighboring C6, the 680µF cap started leaking electrolyte too. I must have accidentally heated up its legs too much. So, looks like the power supply will need a lot of replacement caps, most of which are obsolete or affected by the current component shortage.

Does anybody here have access to a good digital copy of the 1504 service manual? The one from the 1604 has a lot of unreadable parts in it.
Pinörkel is offline  
Old 1st Jan 2022, 6:32 pm   #2
WME_bill
Octode
 
Join Date: Oct 2007
Location: Exeter, Devon, UK.
Posts: 1,553
Default Re: Gould 1504 repair

Pinorkel.
I cannot help with the 1504. But I have posted a complete and good copy of the 1604 manual, which picks up the missing pages (after p68) on the usual web copy. So if that manual could help, there is a better copy available.
Search here on the Forum. My postings were to threads in Nov 2013 by Freya,
Sept 2015 by Willy68 and in Feb 20125 by wme_bill, all under heading Gould 1604 or similar.
wme_bill
WME_bill is offline  
Old 1st Jan 2022, 7:03 pm   #3
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

As you can see in the first image, progress has been made on scope 1. I replaced all the 10µF 40V Philips caps(C14 to C17 and C27) in the PSU, since all of them had a terrible ESR of > 50Ω. One of the small blue caps, the one near the big 5V L296 regulator, is a 2.2 µF 63V one. Do not confuse these, since this one is responsible for setting the soft start duration and the short circuit current. Also, this resistor is missing its label, C9 in the circuit diagram (see green circle in the last picture). The replacements I was able to get are Vishay 021 ASM and do not have a great ESR(~10Ω @ 100Hz), but they seem to work well.

For the two leaky red 220µF caps C10 and C23 and the 680µF cap C6 that I unfortunately overheated during cleaning up the mess of the first two, I swapped in some low ESR replacements from Vishay and Panasonic. The corrosion on C10 and C23 was so strong that it had eaten away their legs internally. Apart from replacing the caps I only had to retouch several solder joints and did some sanity checks on the remaining components. Apparently the resistors are very high quality since they all were spot on.

After that, I put the PSU back into the scope and the 1504 powered on again, passed its self tests and seems to operate normally, except for some glitches. Now, all the voltages apart from the +5V rail were spot on and the PSU internal ripple specs on the voltage regulators were met. However, the +5V rail from the L296 regulator is still sitting too high at 5.4V, while the allowed range is between 4.8V and 5.2V.

To determine the reason for that, I checked the L296 datasheet. For setting the voltage of the regulator, a feedback voltage has to be fed into pin 10 of the regulator and is then compared against the internal 5.1V reference. According to the application circuit diagrams, the feedback voltage is realized with a voltage divider between the regulator output and ground. The resistor of pin 10 to ground should always be 4.7kΩ. The other resistor can then be tuned to set the desired output voltage. Consequently, tying pin 10 directly to the output results in the regulators minimum output voltage of 5.1V. The PSU of the 1504 uses almost exactly the reference circuit, including the 4.7kΩ resistor of pin 10 to ground. However, this one is positioned on the mainboard near the PSU connector and connected via the blue lead, as shown in the fourth image. Not only is it not included in the mainboard layout diagram, I also could not find it in any of the circuit diagrams. The second voltage divider resistor is R18 on the PSU board and it has a value of 680Ω on the PSUs of both scopes. However, the parts list lists it with a value of 100Ω.

Now, it gets really confusing: Neither a value of 680Ω nor a value of 100Ω would theoretically lead to the observed 5.4V output voltage. On both scopes, R18 has exactly the 680Ω, designated by its color code. Additionally, the 4.7kΩ resistor on the mainboard measures only 530Ω on both scopes with the PSU connector disconnected to ensure correct measurement. The trace on the mainboard connecting the blue PSU cable to the 4.7kΩ resistor does not seem to be connected to anywhere else (also checked the internal layers of the PCB by shining a strong light through the PCB), so the measurement should be correct. On the other hand it cannot be correct, since resistances of 680Ω and 530Ω would not lead to the observed 5.1V on the feedback line while the output voltage is nailed to 5.4V.

At the moment, I am kind of lost here. I think it is kind of unlikely that the 4.7kΩ resistors on both scopes degraded to the same value of 530Ω, so I most likely have a measurement error here. Additionally, the fact that the +5V line on both scopes reads exactly 5.4V kind of makes this look intentional. Maybe Gould changed the output voltage without updating the manual? Many of the HC74xxx chips used on the mainboard seem to be operable between 2V and 6V, so maybe 5.4V is totally fine. Has anyone here experiences regarding this or measurements of other 1600 series scopes at hand?
Attached Thumbnails
Click image for larger version

Name:	1504_working.jpg
Views:	166
Size:	166.5 KB
ID:	248783   Click image for larger version

Name:	1504_psu.jpg
Views:	143
Size:	104.3 KB
ID:	248784   Click image for larger version

Name:	1504_psu_replaced_caps.jpg
Views:	129
Size:	42.2 KB
ID:	248785   Click image for larger version

Name:	1504_feedback_resistor.jpg
Views:	137
Size:	95.8 KB
ID:	248786   Click image for larger version

Name:	1504_psu_diag1.png
Views:	131
Size:	22.3 KB
ID:	248801  

Pinörkel is offline  
Old 1st Jan 2022, 7:13 pm   #4
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Quote:
Originally Posted by WME_bill View Post
Pinorkel.
I cannot help with the 1504. But I have posted a complete and good copy of the 1604 manual, which picks up the missing pages (after p68) on the usual web copy. So if that manual could help, there is a better copy available.
Search here on the Forum. My postings were to threads in Nov 2013 by Freya,
Sept 2015 by Willy68 and in Feb 20125 by wme_bill, all under heading Gould 1604 or similar.
wme_bill
Thank you for the hint. I already searched and everything I could find on the 160x scopes on the forum, but only found some service manual fragments, which were of even worse quality than the ones I have. So, I must have missed your posts. I will use your data and search again.

Comparing my 1504 with the 1604 service manual, the only difference seems to be that it has only 10Mpts instead of 20Mpts memory. Other possible differences could well only be options: I already saw 1604s with no graticule illumination pot at the front panel and without a plotter, which my 1504 both do not have.
Pinörkel is offline  
Old 2nd Jan 2022, 9:15 pm   #5
WME_bill
Octode
 
Join Date: Oct 2007
Location: Exeter, Devon, UK.
Posts: 1,553
Default Re: Gould 1504 repair

Pinorkel.
Gould 1604 family.
I'm afraid my earlier posting was not accurate. The full manual is far too large to post on the forum. I am referring to the existence of good copies of the servicing and operating manuals, which I have available and can produce on a CD on request. For they also are too large really to send by e-mail: over 30Mb.
wme_bill
WME_bill is offline  
Old 3rd Jan 2022, 11:57 pm   #6
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Quote:
Originally Posted by WME_bill View Post
Pinorkel.
Gould 1604 family.
I'm afraid my earlier posting was not accurate. The full manual is far too large to post on the forum. I am referring to the existence of good copies of the servicing and operating manuals, which I have available and can produce on a CD on request. For they also are too large really to send by e-mail: over 30Mb.
wme_bill
Hello wme_bill,
thank you for the kind offer. A good quality service and operating manual for the 1600 series would be very nice. If you cannot send 30MB files via email, I could provide you via PM with a secure password protected link to a fast upload service(upload limit > 2GB), where multiple large files can be uploaded simply via drag & drop in the browser.

Denis
Pinörkel is offline  
Old 5th Jan 2022, 8:20 pm   #7
G6fylneil
Pentode
 
Join Date: Sep 2011
Location: Coventry, West Midlands, UK.
Posts: 211
Default Re: Gould 1504 repair

I discovered the same 5V rail sense problem with my 1604. I assumed the PSU had been changed at some time, but I didn't have another 1604 to compare it to. I lifted one end of the resistor (which isn't on the component layout I have), which I assumed was half of a potential divider and jumpered the sense line to the 5V rail. I think my 1604 was built in 1990, I bought it in 1999, and it worked well for many years, but now the little tants keep blowing, the front panel push switches have gone high resistance, and the fumes from the mempaks have stopped the RTC from working, but I don't have the enthusiasm to fix it at the moment.
G6fylneil is offline  
Old 6th Jan 2022, 8:00 pm   #8
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Quote:
Originally Posted by G6fylneil View Post
I discovered the same 5V rail sense problem with my 1604. I assumed the PSU had been changed at some time, but I didn't have another 1604 to compare it to. I lifted one end of the resistor (which isn't on the component layout I have), which I assumed was half of a potential divider and jumpered the sense line to the 5V rail. I think my 1604 was built in 1990, I bought it in 1999, and it worked well for many years, but now the little tants keep blowing, the front panel push switches have gone high resistance, and the fumes from the mempaks have stopped the RTC from working, but I don't have the enthusiasm to fix it at the moment.
The 5V line being at 5.4V seems to be somewhat common. I also found one repair at the eevblog forum that dealt with this by simply routing the sense line to the output voltage. What bugs me most regarding this is that the used resistors would not result in 5.4 volts according to the regulator datasheet and the measured values of the resistor on the mainboard differ even more from the theoretically intended values.

Something useful for all who have a unit with a plotter and cannot find pen replacements: Apparently this kind of pens work and are still available.
Pinörkel is offline  
Old 7th Jan 2022, 11:56 am   #9
G6fylneil
Pentode
 
Join Date: Sep 2011
Location: Coventry, West Midlands, UK.
Posts: 211
Default Re: Gould 1504 repair

My 1604 did manage to burn out one of the little yellow 100uH chokes in the horizontal circuitry, I assumed the supply sequencing went wrong and the chip it fed "crowbarred" but all was well when I replaced it.
G6fylneil is offline  
Old 8th Jan 2022, 12:04 am   #10
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

I was not able to solve the mystery of the 5V regulator sense line yet. The only other issue, for now, was an occasional startup issue, where the 5V line somehow fails to start up. Switching the scope off and on again immediately solves this. In this failure state, the 5V regulator only outputs about 0.4V, although the input voltage and ripple at pin 3 of the regulator seems to be OK. I could not find the source of this yet. Maybe a faulty L296 regulator?

However, since the scope seemed to work almost completely, I started experimenting on the interface of the internal plotter that my unit is not equipped with. I managed to fake the presence of the plotter by pulling one pin to ground and reading serial output with an ESP32. There, I still have to solve some logic level conversion issues.

Then, during one of the scopes recalibration phases, which occur about every 10 minutes or so, the scope unexpectedly rebooted. Everything was working normally after that again. No smoke, no strange smell, nothing. A short while later, the scope got stuck in a repeated reboot hiccup until it finally set all vertical ranges to 500V and printed "ERROR !" at the bottom of the screen. At this point, the controls were not functional anymore. Switching it of and on again brought back normal operation. Tapping the device for loose contacts was inconclusive and I could not reproduce the issue for simultaneously having a look at the PSU output voltages. So I switched the device off to do some measurements on the PSU components, but found no clues. I think, the displayed error was caused by memory corruption from the boot loop hiccup, which may have been caused by some kind of PSU failure. My main suspect is the L296 regulator for the 5V rail.

So I attached my logging DMM to the 5V output, switched the scope back on and nothing happened. No lights, no fan, no AC voltages on the PSU inputs. Turned out, the fuse had silently blown. Great! Now I am stuck with no fuse, since the last electronics shop around here closed a short while ago and my usual online sources cannot deliver fuses with the right specs at the moment. So continuing work on the scope is out of question until I get a supply of fitting fuses.

Maybe the fuse blew as a result of the boot loop hiccup. On the other hand it may also be connected with the voltage selector at the back of the scope still being on 220V instead on 240V. I did not switch this because with the 220V setting all voltages in the PSU were much closer to their reference values than on the 240V setting.
Attached Thumbnails
Click image for larger version

Name:	1504_error.jpg
Views:	64
Size:	80.8 KB
ID:	249258  
Pinörkel is offline  
Old 10th Jan 2022, 1:16 am   #11
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

The lack of fuses gave me time to investigate the mysterious 5.4V output on the 5V line. Turns out, 5.4V is the value Gould intended by design that most likely should not be lowered by bypassing the feedback circuit. To substantiate this, I tried to isolate the responsible circuit. The first picture shows the reference circuit for the L296 regulator. Note the variable voltage divider at the right side to the feedback pin 10 with the 4.7kΩ resistor to ground. On the Gould 1600 series, although listed in the PSU parts list and not shown in the schematics, this resistor is located on the mainboard near the power connector. Only after removing it, the label R30 from the parts list is visible, as can be seen in the second image. However, there is a second 470Ω resistor in the parts list, labeled R31, which is not show in the schematics either. Since the traces leading there are 100% invisible, it took me a lot of tapping around with a piece of aluminum foil connected to my DMM to locate this one on the mainboard. It is hidden in the middle of the board below the acquisition board besides the large 470µF capacitor as can be seen in the third and fourth image. Its other side is connected to a network of wide inner layer traces that form the 5V supply line for many of the ICs. All in all, the sense circuit looks like the schematic in the last image. Initially, R18 with a value of 680Ω on the psu board and R30 with a value of 4.7kΩ seem to form the usual voltage divider of the L296. The reason for this not making sense at all when measuring the resistor values and calculating the theoretical resulting output of the L296 is the hidden R31 with a value of 470Ω on the mainboard which is connected in parallel with R18. Now everything makes sense and an output value of 5.4V of the L296 matches perfectly with the resulting sense voltage of 5.1V. All resistor values were spot-on. So, why has Gould raised the voltage to 5.4V instead of the allowed 4.8V to 5.2V according to the specs without updating the specs? Maybe they updated the specs and the only available service manual is an older one. Maybe they tried to increase stability of the digital circuits. Who knows?
Attached Thumbnails
Click image for larger version

Name:	l296_reference_circuit.png
Views:	99
Size:	33.8 KB
ID:	249417   Click image for larger version

Name:	1504_psu_r30.jpg
Views:	96
Size:	95.6 KB
ID:	249418   Click image for larger version

Name:	1504_psu_r31_top.jpg
Views:	93
Size:	128.6 KB
ID:	249419   Click image for larger version

Name:	1504_psu_r31_bottom.jpg
Views:	91
Size:	116.0 KB
ID:	249420   Click image for larger version

Name:	psu_feedback_circuit.png
Views:	82
Size:	11.9 KB
ID:	249421  

Pinörkel is offline  
Old 10th Jan 2022, 8:05 pm   #12
G6fylneil
Pentode
 
Join Date: Sep 2011
Location: Coventry, West Midlands, UK.
Posts: 211
Default Re: Gould 1504 repair

My copy of the service manual doesn't have an issue or version No. but on the back page the last line starts "WL 7 87..." which I always took to be somebody's initials and a date.

I agree with your analysis of the 5V remote sense circuit.

On some systems I used to maintain, the 5V supplies were set high because there were Schottkys in the supply rails.

At a company I used to work at the customer service manuals were contracted-out to specialist technical authors. We engineers used the factory drawings from the drawing office.

When the yellow choke burnt out I did suspect a high resistance connection at the IDC connector from the PSU. I have had similar Molex KK connectors go high resistance (to the board mounted pin) on PSUs, sometimes discolouring the housings, but I don't know if the AMP connectors suffer the same problem.

When I successfully located and replaced the first short circuit tant. I refitted all the covers, and their screws, switched it on, it went through the cal. and then the caps. in the Shaffner mains filter failed, fortunately the plastic boot around it contained the majority of the "mess".
G6fylneil is offline  
Old 15th Jan 2022, 8:58 pm   #13
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Quote:
Originally Posted by G6fylneil View Post
When the yellow choke burnt out I did suspect a high resistance connection at the IDC connector from the PSU. I have had similar Molex KK connectors go high resistance (to the board mounted pin) on PSUs, sometimes discolouring the housings, but I don't know if the AMP connectors suffer the same problem.
I did not know that these are susceptible to burning out.
Quote:
Originally Posted by G6fylneil View Post
When I successfully located and replaced the first short circuit tant. I refitted all the covers, and their screws, switched it on, it went through the cal. and then the caps. in the Shaffner mains filter failed, fortunately the plastic boot around it contained the majority of the "mess".
I already read that these have a tendency to fail. Is that due to some exploding Rifa caps, evaporating the potting compound?

In the meantime, I continued fault analysis on the intermittently failing 5V rail. Yesterday the scope ran for hours with my Keithley 6500 attached, but the fault did not show up. A heisenbug maybe? Finally today, I could capture it several times while monitoring the 5V output and the input of the L296. As can be seen in the first image, there is no power loss on the input, just a little spike, as if the regulator had just turned off. On the output, which is very stable in working condition, there is a voltage drop to about 0.4V, then a spike followed by an even deeper voltage drop and then it goes back to normal. The 0.4V got me thinking, because I remembered the L296 datasheet where it states:
Quote:
Output overload protection is provided in the form of a current limiter. The load current is sensed by an internal metal resistor connected to a comparator. When the load current exceeds a preset threshold this comparator sets a flip flop which disables the output stage and discharges the soft start capacitor. A second comparator resets the flip flop when the voltage across the soft start capacitor has fallen to 0.4V. The output stage is thus re-enabled and the output voltage rises under control of the soft start network. If the overload condition is still present the limiter will trigger again when the threshold current is reached. The average short circuit current is limited to a safe value by the dead time introduced by the soft start network.
So, apparently there is something going on that resembles the overload protection handling behavior of the L296. Maybe the L296 is not defective, but there is an intermittent short circuit on the board, triggering the overload handling of the L296. Unfortunately I have no parts for an adapter to put between the PSU and the mainboard with my Keithley 6500 in current mode in the 5V line. Maybe I can borrow a thermal camera and try to find something on the mainboard. At least when the fault condition occurs repeatedly there should be a component that gets warm enough to show up in a thermal image.

Another thing, I discovered is that there is a transient voltage suppressor diode (D9) between 5V and ground on the mainboard near the power connector. From what I could read it is an ST 1N5908. This one has a stand-off voltage of 5V and a breakdown voltage of 6V. So, Gould intentionally raising the 5V line to 5.4V most likely puts more stress on that diode than necessary. I will have to keep an eye on that, although I found damage-free desoldering components on the multi layer mainboard to be very difficult without a proper desoldering iron. But good quality ones e.g. from JBC are quite expensive and would take a lot of space on my non-existing electronics bench.
Attached Thumbnails
Click image for larger version

Name:	5V_power_loss_L296_input.jpg
Views:	39
Size:	29.6 KB
ID:	249746   Click image for larger version

Name:	5V_power_loss_L296_output.jpg
Views:	34
Size:	28.8 KB
ID:	249747  
Pinörkel is offline  
Old 19th Jan 2022, 1:45 am   #14
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Some progress was made. I decided to use a divide-and-conquer strategy to locate the fault. So I substituted the mainboard on the 5V line by a 6V incandescent lamp from an old bicycle light, switched the PSU on, and waited for the fault to show up. It took quite some time, but luckily it showed up. So I knew, the fault was located in the 5V part of the PSU and most likely not on the mainboard. From an earlier over-current test with an external lab power supply I also knew that the mainboard draws ~850 mA from the 5V line, when no other power lines are connected. The incandescent lamp draws only 400 mA and it took longer for the fault to occur. That, and the fact that the fault occurred sooner when I loaded the 5V line with an additional 60 mA made me suspect a temperature issue. The only hot component in the 5V part of the PSU was the L296 regulator, but the service manual states that it is expected to do so. Since the freewheeling diode had been tackled by capacitor fumes and seen some heat during my clean-up attempts, I changed it for a new MUR415. This caused the issue to occur less frequently, so maybe it was part of the issue.

Then I realized that I had done all my testing with the top cover off. The only fan in the case blows cold air inside, but if the top cover is not on, this air is not forced to exit the case through the ventilation channels near the PSU, but just exits through the open top side. To test this, I put the top cover on and the scope ran for hours without restarting or showing any signs of a 5V power loss.

The scope seems to be running fine now. However, I still do not trust its current state. According to its specification the L296 can supply up to 4 A of current and the rating of the 5V line in the 1600 service manual looks like 2.5 A. So, there should be plenty of headroom between the 850mA power consumption I measured for the 5V line with no other power connected to the 2.5 A. Removing the top cover or adding 60 mA power consumption on the 5V line should be no issue at all. Or did Gould really design the scope with such tight thermal tolerances?
Pinörkel is offline  
Old 21st Jan 2022, 6:59 pm   #15
G6fylneil
Pentode
 
Join Date: Sep 2011
Location: Coventry, West Midlands, UK.
Posts: 211
Default Re: Gould 1504 repair

I would be cautious operating the scope with just the 5V supply and then applying the 15V rails. I worked on some systems in the 1990s with big linear power supplies, (80A 5V and 30A 15V and parallel redundancy!) and if you turned the power off and on quickly with most of the system boards missing, you would get a pop-pop-pop noise as some AD(7xxx?) analog switches blew up!
G6fylneil is offline  
Old 21st Jan 2022, 7:51 pm   #16
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Quote:
Originally Posted by G6fylneil View Post
I would be cautious operating the scope with just the 5V supply and then applying the 15V rails. I worked on some systems in the 1990s with big linear power supplies, (80A 5V and 30A 15V and parallel redundancy!) and if you turned the power off and on quickly with most of the system boards missing, you would get a pop-pop-pop noise as some AD(7xxx?) analog switches blew up!
Thank you for the hint. I performed the test with an external +5V power supply, current limited to 2.5A, because the service manual explicitly recommends to do this on page 19, to check the scope for an over-current situation on the mainboard. For this test, only +5V are supplied to the mainboard and all other rails remain unpowered.
Pinörkel is offline  
Old 19th Feb 2022, 12:48 am   #17
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Still on it. After ruling out as many other options as possible, I decided to try replacing the L296 regulator. Without a desoldering iron with a vacuum pump this was a nightmare. Doing this with hot air would have damaged too many other components on the board, so I had to suck all the legs as dry as possible using my Engineer SS-02 solder sucker and solder wick. The square legs were such a tight fit, that wiggling the regulator free damaged quite some inner platings of the through holes. (note to self: finally get a decent jbc desoldering station for this)

After extracting the regulator, I screwed it to a heat sink, build the minimal 5.1V regulator circuit from the L296 datasheet using some parts from my rescued parts box and got myself a set of 50W power transistors for testing. This revealed that the regulator was indeed quite unstable, supporting only light loads and also being quite sensitive to changes of the input voltage. Since I could also observed the constant restart behavior, I was confident enough to solder the new regulator into the PSU. With this, the scope runs without any flaws for hours. The mean deviation on the 5.4V supply voltage is minimal (< 1mV after two hours) and mostly comes from a thermal drift of internal L296 the reference voltage.

As before, the regulator heatsink gets moderately hot (~50°C). Closing the top cover of the scope brings this down considerably by about 10°C. Now I need to find a damage free way to verify, if the +5V rail will support an additional 150mA of load for a small plotting related gadget I want to integrate into the scope. In theory, the L296 should support up to 4A load and the Gould PSUs overload reference value is at 2.5A. Quite some room from the 0.8A power consumption, I measured with an external PSU. However, this can only work, if the thermal design provides enough cooling for the L296, which I am quite unsure it does.

In addition to the repair, I noticed a difference on the 1504 main board to the available 1604 service manual. Near the back side of the scope there is an undocumented 34 pin port header (see image). The 1604 seems to have a NAND, a NOR chip and a capacitor in that place instead. The connector has the same size as a floppy drive connector but measurements suggest it is some kind of parallel interface. Maybe the interface for the thermal plotter option for continuous plotting?
Attached Thumbnails
Click image for larger version

Name:	1504_undocumented_port.jpg
Views:	75
Size:	86.5 KB
ID:	252152  
Pinörkel is offline  
Old 19th Feb 2022, 11:29 am   #18
dave cox
Nonode
 
dave cox's Avatar
 
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,059
Default Re: Gould 1504 repair

Alternative, you could derive another 5V supply from the existing regulators input and use a buck converter to get 5V without producing much heat. Tricky to do without introducing noise, but not impossible if you have the room ...

dc
dave cox is offline  
Old 19th Feb 2022, 1:28 pm   #19
mole42uk
Nonode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Resolfen, Wales; and Bristol, England
Posts: 2,588
Default Re: Gould 1504 repair

Quote:
Originally Posted by Pinörkel View Post
Something useful for all who have a unit with a plotter and cannot find pen replacements: Apparently this kind of pens work and are still available.
Yes, I have purchased pens from that site an can confirm they work in the ALPS printers on Gould 'scopes. I have a 1604 and a 4074.....
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is online now  
Old 20th Feb 2022, 3:02 pm   #20
Pinörkel
Pentode
 
Join Date: Jun 2019
Location: Dortmund, Germany
Posts: 161
Default Re: Gould 1504 repair

Quote:
Originally Posted by dave cox View Post
Alternative, you could derive another 5V supply from the existing regulators input and use a buck converter to get 5V without producing much heat. Tricky to do without introducing noise, but not impossible if you have the room ...
dc
Unfortunately, my gimmick relies on being fed the same supply voltage as the logic level of the 1504. I could solve that issue by using a logic level converter, but I want to keep the number of required parts and the wiring complexity as low as possible.


Quote:
Originally Posted by mole42uk View Post
Yes, I have purchased pens from that site an can confirm they work in the ALPS printers on Gould 'scopes. I have a 1604 and a 4074.....
Thank you for the confirmation. I just noticed the link is no longer valid. Here is an updated one.
Pinörkel is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 5:40 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.