Gould DSO 4094 SVG DEAD Status - Page 2 - UK Vintage Radio Repair and Restoration Discussion Forum

PBERTRAM · 10th Apr 2020, 8:34 pm

I've taken pictures of plugged ADC board and both CCD boards.
On ADC board, there are 2 RF IN and 2 RF OUT connectors going to and from each CCD boards.
Is timebase board the one that is between 2nd CCD board and ADC board ?
I can see also RF cable directly soldered on CCD boards -> is that for calibration ?
One thing I noticed is that CCD boards are differents, maybe one of them has been already changed. Moreover, one has a red tuning and the other a white one, and those selector do not have same position value. What does this tuning correspond to ?

Noopy2014 · 10th Apr 2020, 8:45 pm

Sorry, there was a mistake on my side: The 4094 has only one calibration clock cable.
The digital Timebase board is the one between adc and ccd.
The calibration cable is the one soldered to the board, yes.

The selectors somehow delay the signal so that die signal arrives at the ccd in the right Moment. They have to show different numbers.

PBERTRAM · 11th Apr 2020, 7:21 am

Just tried to check the voltages on all RF soldered cable. Nothing when my multimeter is in AC, and about -1.75V in DC. I presume this is because frequency is too high for AC voltage to be measured this way. The only thing positive is that all voltages are about the same in DC. In fact, I would need... a scope to investigate here.
Another thing I noticed is the jump that are near the CCD selectors : they are not at the same place on each board. I have JM2 in place for CCD board selector on 3 and JM3 for CCD board selector on 2. But I presume this is normal ?

Noopy2014 · 11th Apr 2020, 8:12 am

I'm afraid without an oscilloscope it will be a long way to find the failure...
The Jumpers belong to the delay circuit. You shouldn't change them. If there is a problem you only have problems with the exact timing of the sampling points at some timebases. No problem right now.

I still assume there is a communication problem somewhere... Perhaps one of the logic chips connecting the different blocks to the controller...

Let's look once more at the startup:

"ROM TEST,
CH1 AC, I heard several relays click and I think it is just before
CH1 DC and immediately after
YSH INIT,
CAL YSH (very fast),
OFFSET SCL,
GAIN SCL,
08 CAL,
G8 CAL (it goes back and forth several times between 08 CAL and G8 CAL), and then
RF CAL (stuck here)."

ROM seems good. That's good.
It sounds like the input amplifier path is good (CH1 AC to probably GAIN SCL). That's very good because it's completely different than one in the 4074.
I'm pretty sure that O8 CAL corrects the offset of a CCD and G8 CAL corrects the gain of a CCD. Every CCD has eight lines that have to be corrected. That also explains why the 4094 switches between both tasks (4 CCDs). These correction circuits are NOT on the CCD boards.
I assume RF CAL is the calibration of the CCD timing. The eight lines have to sample at the right time an that is calibrated. This calibration is on the CCD board. The signal comes from the digital timebase board. If there is a calibration signal then the next possible candidate is the logic on the CCD board.
The controller talks over it's bus, selects the necessary parts on the CCD board and then delivers a analogue signal over the Input/Output board (the last one).
The analogue part is probably ok. You can't be sure but it also is necessary for the input amplifier calibration. You have a lot of logic on the CCD board that converts the bus signals to control signals switching different parts for sampling calibration. If there is a small logic dead the controller can't activate the necessary parts.
A failed part in the analogue part seems unlikely because the gain and offset calibration was ok.

Puh... I won't read that text again. I hope my english doesn't make your brains bleed...

PBERTRAM · 11th Apr 2020, 11:28 am

Finally, I tried to press "Abort" button BEFORE starting the scope.
And that changes ALL !
It display error message on screen and "CH4 TV-F" on status display (see Pic 1). Many leds are lit also and most important is Master Menu is available now (see Pic 2).
Because of TV-F message on display, I quickly pressed TV and Special functions button that leads to another menu (see Pic 3). It is possible to start a partial calibration here, so that's what I tried. After a while, it finally passed it well.
In fact, first trigger cal seems to be difficult, so I managed to go through it which first leads to unstable display. After checking all but the trigger was working well, I started another partial cal that finally leads to the results in Pic 4.
This scope seems to work well as far as partial calibration is performed. All channel work, which is even more than I need ! And Auto Setup seems very convenient to tune immediatly the scope to the signal.
Now I'm just curious to understand what TV functions are for (I presume it is linked to RF CAL) ?

mole42uk · 11th Apr 2020, 1:38 pm

Quote:

Originally Posted by PBERTRAM

Finally, I tried to press "Abort" button BEFORE starting the scope.
And that changes ALL !

That's the reason I suggested you take that procedure in post number 8....

Noopy2014 · 11th Apr 2020, 2:00 pm

Interesting...
The TV-F is no error. The Dot-Matrix-Display shows the chosen Trigger. The TV-F stands for TV-Trigger.
Good to hear that the 4094 does it's job now but I wouldn't bet on it since it seems there are hiding one or more problems inside.

PBERTRAM · 11th Apr 2020, 6:08 pm

Yes mole42uk, you're right. The pb is I did not press "Abort" button on the right timing.
I find the idea good and tried it again with success this time. I must thank you very much for that!
In fact, I tried several times to understand Abort button has to be pressed only once during ROM TEST. Then it displays GOODBYE and you can have access to all command and control.
Operating this way (without cal) seems possible but signal is noisy.
But I confirm the trigger problem : trigger cal is abnormally long during partial cal.
And it is even worst during complete cal with RF signal. Maybe it could pass if I wait more because it finally did the very first time I started this scope.
So my problem has to do with trigger for sure, all other functions seems OK.
Do you know where I should check to try to find the origin of this problem ? If I don't find it, you're probably right Noopy concerning reliability of this scope : it will soon be barelly usable...
I thank you all for your help guys, I would never have progressed without all your advices and suggestions for sure !

Noopy2014 · 11th Apr 2020, 6:48 pm

If you have a problem with the trigger circuit itself the error traking will get more complicated. I'm the 407x and the 408x the trigger circuit is on a small board in front of the large boards. In the 4094 with the switch to 200MHz Gould had to change the trigger circuit completely and moved it to the boards under the cathode ray tube. I don't know anything about that circuit. Perhaps you can find dead components by looking very close or you can find loose connections or something like that. Of course some reverse engineering would be best...

You can also check if there is a special trigger function that doesn't work properly. Perhaps that helps...

PBERTRAM · 12th Apr 2020, 10:33 am

I progressed a bit on understanding how this DSO works.
After Cal abortion at start-up, I went in the 2nd page of Master Menu and found a "Manual Setup Menu" in it. Choosing "8 phase & Y-shift Calibration" allows me to improve the signal display (see attached pictures before and after manual calibration). I get rid of some noise and signal stability is just fine operating this way. I also found channel 2 does not seems to work (I certainly made a bad check on this point).
Finally I suspect this scope has been throwed away because it is obsolete with breakdown on channel 2. Therefore it would essentially need a good overall cal. I measured about 0.5V on front panel Cal 1V pin. How could I manage that ? And what about the tunable resistors on timebase and CCD boards (R229, R225, etc...) ?
Thanks again for your precious help !

Malcolm G6ANZ · 12th Apr 2020, 11:24 am

This may be of some use. I found it on the same site as the die photos. Its in German but Google coud be persuaded to translate for you.
https://richis-lab.de/Gould407X.htm

mole42uk · 12th Apr 2020, 11:41 am

Quote:

Originally Posted by Malcolm G6ANZ

This may be of some use. I found it on the same site as the die photos. Its in German but Google coud be persuaded to translate for you.
https://richis-lab.de/Gould407X.htm

I've already translated about half of that....

Noopy2014 · 12th Apr 2020, 11:53 am

You can also ask me directly...

Some time I will also reverse engineere the 4094. Right know review status of the 4074 teardown is page 64 of 389. Give me some more years...

As you can see I own some Gould 40xx...

PBERTRAM · 12th Apr 2020, 12:45 pm

Sure I will. You already provide me with interesting data concerning power-supply voltage tuning, so maybe you have some more concerning CCD and timebase board as well as channel amplifiers (and CAL 1V) tuning ?
I don't think a complete service manual or scope design analysis is necessary for doing this. But starting from nothing and see what happen when I tune something on those board may not be the best method I think...

Noopy2014 · 12th Apr 2020, 1:06 pm

I have round about 400 pages to tell...

Where do you want me to start?

The 1V calibration clock is generated on the input output board (the one beside the CRT. R37 sets the level.

I would suggest to do some more investigation where the 4094 has problems. Just turning some screws will probably worsen the problems.

What are the errors right now?
- Calibration still stops at startup?
I would try to fix this one first if possible.
- Ch2 is dead?
I would first check whether the signal is available at the CCD board connector.

PBERTRAM · 12th Apr 2020, 3:39 pm

I check "CAL 1V" pin signal (see corresponding picture). Not too bad...
Then I found a Test Menu that allows me to select single calibrations. Interesting to understand better how the scope works step by step. I did all but trigger cal that is long and can put the mess. I could see that each time a calibration involves trigger function (such as GAIN_CAL, see pic with front panel red led blink), the corresponding cal takes time. It seems the scope has to try several times before succeeding to pass.
Finally I displayed the Cal status that shows a problem on channel 2 concerning both maximum gain and amplifier corrections. But variable gain is OK even for channel 2. What does that mean ?
I think I can live with 3 channels only. For me, trigger cal is my biggest pb on this scope. Nevertheless it still remains usable as far as I perform selected cals the way I did...

Noopy2014 · 12th Apr 2020, 4:54 pm

Unfortunately we don't know very much about the input and the trigger circuit of the 4094.

Did you already check the boards under the CRT optically? Perhaps you find some strange looking things.
You can also try to find supply voltage lines and measure them. Perhaps there are dead tantalum capacitors somewhere.
You will have to do some reverse engineering. It's a pity we don't have a service manual for the 4094..

PBERTRAM · 12th Apr 2020, 8:06 pm

Sure I took a look at all, since I remove both top and bottom cover and unplugged boards to check all connections... without success. All seem OK.
But again, trigger and auto-setup works well in normal operation, the issue is always with CAL involving trigger circuit. Even if 4074 and 4094 scope have not the same circuits, having info on 4074 trigger cal circuit may be helpful to understand how it works.

Noopy2014 · 12th Apr 2020, 8:24 pm

The 4074 and the 4094 are very different regarding the trigger.
In the 4074 the trigger board has no calibration options.
On the input amplifier board you can adjust the offset of the two signals going to the trigger board (CH1 and CH3).
That's all.

From the Input/Output board two analog signals are reaching the trigger board and control the trigger level.
A serial bus controls the trigger board. A serial to parallel converter generates the signals that activate different couplings, low pass and high pass and TV functions.

The only self calibration function that I can think of is a calibration of the trigger level.

The trigger board is controlled by a MM5451. Perhaps the 4094 has also a MM5451. You can try to check whether that is still ok. => communication

PBERTRAM · 13th Apr 2020, 2:52 pm

Yes you're right (again), there is one MM5451 for the specific TV trigger circuit. But in fact there are many others on the board !
4094 is just 4074 evolution and Gould certainly reused a lot from 4074 for the new one.
But the more I have a look on the boards, the less I see burned device or whatever. Overall is rather clean with even very few suspect soldering.

10th Apr 2020, 8:34 pm	#21
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status I've taken pictures of plugged ADC board and both CCD boards. On ADC board, there are 2 RF IN and 2 RF OUT connectors going to and from each CCD boards. Is timebase board the one that is between 2nd CCD board and ADC board ? I can see also RF cable directly soldered on CCD boards -> is that for calibration ? One thing I noticed is that CCD boards are differents, maybe one of them has been already changed. Moreover, one has a red tuning and the other a white one, and those selector do not have same position value. What does this tuning correspond to ? Attached Thumbnails

10th Apr 2020, 8:45 pm	#22
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status Sorry, there was a mistake on my side: The 4094 has only one calibration clock cable. The digital Timebase board is the one between adc and ccd. The calibration cable is the one soldered to the board, yes. The selectors somehow delay the signal so that die signal arrives at the ccd in the right Moment. They have to show different numbers. __________________ https://www.richis-lab.de

11th Apr 2020, 7:21 am	#23
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status Just tried to check the voltages on all RF soldered cable. Nothing when my multimeter is in AC, and about -1.75V in DC. I presume this is because frequency is too high for AC voltage to be measured this way. The only thing positive is that all voltages are about the same in DC. In fact, I would need... a scope to investigate here. Another thing I noticed is the jump that are near the CCD selectors : they are not at the same place on each board. I have JM2 in place for CCD board selector on 3 and JM3 for CCD board selector on 2. But I presume this is normal ?

11th Apr 2020, 8:12 am	#24
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status I'm afraid without an oscilloscope it will be a long way to find the failure... The Jumpers belong to the delay circuit. You shouldn't change them. If there is a problem you only have problems with the exact timing of the sampling points at some timebases. No problem right now. I still assume there is a communication problem somewhere... Perhaps one of the logic chips connecting the different blocks to the controller... Let's look once more at the startup: "ROM TEST, CH1 AC, I heard several relays click and I think it is just before CH1 DC and immediately after YSH INIT, CAL YSH (very fast), OFFSET SCL, GAIN SCL, 08 CAL, G8 CAL (it goes back and forth several times between 08 CAL and G8 CAL), and then RF CAL (stuck here)." ROM seems good. That's good. It sounds like the input amplifier path is good (CH1 AC to probably GAIN SCL). That's very good because it's completely different than one in the 4074. I'm pretty sure that O8 CAL corrects the offset of a CCD and G8 CAL corrects the gain of a CCD. Every CCD has eight lines that have to be corrected. That also explains why the 4094 switches between both tasks (4 CCDs). These correction circuits are NOT on the CCD boards. I assume RF CAL is the calibration of the CCD timing. The eight lines have to sample at the right time an that is calibrated. This calibration is on the CCD board. The signal comes from the digital timebase board. If there is a calibration signal then the next possible candidate is the logic on the CCD board. The controller talks over it's bus, selects the necessary parts on the CCD board and then delivers a analogue signal over the Input/Output board (the last one). The analogue part is probably ok. You can't be sure but it also is necessary for the input amplifier calibration. You have a lot of logic on the CCD board that converts the bus signals to control signals switching different parts for sampling calibration. If there is a small logic dead the controller can't activate the necessary parts. A failed part in the analogue part seems unlikely because the gain and offset calibration was ok. Puh... I won't read that text again. I hope my english doesn't make your brains bleed... __________________ https://www.richis-lab.de

11th Apr 2020, 11:28 am	#25
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status Finally, I tried to press "Abort" button BEFORE starting the scope. And that changes ALL ! It display error message on screen and "CH4 TV-F" on status display (see Pic 1). Many leds are lit also and most important is Master Menu is available now (see Pic 2). Because of TV-F message on display, I quickly pressed TV and Special functions button that leads to another menu (see Pic 3). It is possible to start a partial calibration here, so that's what I tried. After a while, it finally passed it well. In fact, first trigger cal seems to be difficult, so I managed to go through it which first leads to unstable display. After checking all but the trigger was working well, I started another partial cal that finally leads to the results in Pic 4. This scope seems to work well as far as partial calibration is performed. All channel work, which is even more than I need ! And Auto Setup seems very convenient to tune immediatly the scope to the signal. Now I'm just curious to understand what TV functions are for (I presume it is linked to RF CAL) ? Attached Thumbnails

11th Apr 2020, 2:00 pm	#27
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status Interesting... The TV-F is no error. The Dot-Matrix-Display shows the chosen Trigger. The TV-F stands for TV-Trigger. Good to hear that the 4094 does it's job now but I wouldn't bet on it since it seems there are hiding one or more problems inside. __________________ https://www.richis-lab.de

11th Apr 2020, 6:08 pm	#28
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status Yes mole42uk, you're right. The pb is I did not press "Abort" button on the right timing. I find the idea good and tried it again with success this time. I must thank you very much for that! In fact, I tried several times to understand Abort button has to be pressed only once during ROM TEST. Then it displays GOODBYE and you can have access to all command and control. Operating this way (without cal) seems possible but signal is noisy. But I confirm the trigger problem : trigger cal is abnormally long during partial cal. And it is even worst during complete cal with RF signal. Maybe it could pass if I wait more because it finally did the very first time I started this scope. So my problem has to do with trigger for sure, all other functions seems OK. Do you know where I should check to try to find the origin of this problem ? If I don't find it, you're probably right Noopy concerning reliability of this scope : it will soon be barelly usable... I thank you all for your help guys, I would never have progressed without all your advices and suggestions for sure !

11th Apr 2020, 6:48 pm	#29
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status If you have a problem with the trigger circuit itself the error traking will get more complicated. I'm the 407x and the 408x the trigger circuit is on a small board in front of the large boards. In the 4094 with the switch to 200MHz Gould had to change the trigger circuit completely and moved it to the boards under the cathode ray tube. I don't know anything about that circuit. Perhaps you can find dead components by looking very close or you can find loose connections or something like that. Of course some reverse engineering would be best... You can also check if there is a special trigger function that doesn't work properly. Perhaps that helps... __________________ https://www.richis-lab.de

12th Apr 2020, 10:33 am	#30
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status I progressed a bit on understanding how this DSO works. After Cal abortion at start-up, I went in the 2nd page of Master Menu and found a "Manual Setup Menu" in it. Choosing "8 phase & Y-shift Calibration" allows me to improve the signal display (see attached pictures before and after manual calibration). I get rid of some noise and signal stability is just fine operating this way. I also found channel 2 does not seems to work (I certainly made a bad check on this point). Finally I suspect this scope has been throwed away because it is obsolete with breakdown on channel 2. Therefore it would essentially need a good overall cal. I measured about 0.5V on front panel Cal 1V pin. How could I manage that ? And what about the tunable resistors on timebase and CCD boards (R229, R225, etc...) ? Thanks again for your precious help ! Attached Thumbnails

12th Apr 2020, 11:24 am	#31
Malcolm G6ANZ Octode Join Date: Feb 2006 Location: Bristol, UK. Posts: 1,042	Re: Gould DSO 4094 SVG DEAD Status This may be of some use. I found it on the same site as the die photos. Its in German but Google coud be persuaded to translate for you. https://richis-lab.de/Gould407X.htm

12th Apr 2020, 11:53 am	#33
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status You can also ask me directly... Some time I will also reverse engineere the 4094. Right know review status of the 4074 teardown is page 64 of 389. Give me some more years... As you can see I own some Gould 40xx... __________________ https://www.richis-lab.de

12th Apr 2020, 12:45 pm	#34
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status Sure I will. You already provide me with interesting data concerning power-supply voltage tuning, so maybe you have some more concerning CCD and timebase board as well as channel amplifiers (and CAL 1V) tuning ? I don't think a complete service manual or scope design analysis is necessary for doing this. But starting from nothing and see what happen when I tune something on those board may not be the best method I think...

12th Apr 2020, 1:06 pm	#35
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status I have round about 400 pages to tell... Where do you want me to start? The 1V calibration clock is generated on the input output board (the one beside the CRT. R37 sets the level. I would suggest to do some more investigation where the 4094 has problems. Just turning some screws will probably worsen the problems. What are the errors right now? - Calibration still stops at startup? I would try to fix this one first if possible. - Ch2 is dead? I would first check whether the signal is available at the CCD board connector. __________________ https://www.richis-lab.de

12th Apr 2020, 3:39 pm	#36
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status I check "CAL 1V" pin signal (see corresponding picture). Not too bad... Then I found a Test Menu that allows me to select single calibrations. Interesting to understand better how the scope works step by step. I did all but trigger cal that is long and can put the mess. I could see that each time a calibration involves trigger function (such as GAIN_CAL, see pic with front panel red led blink), the corresponding cal takes time. It seems the scope has to try several times before succeeding to pass. Finally I displayed the Cal status that shows a problem on channel 2 concerning both maximum gain and amplifier corrections. But variable gain is OK even for channel 2. What does that mean ? I think I can live with 3 channels only. For me, trigger cal is my biggest pb on this scope. Nevertheless it still remains usable as far as I perform selected cals the way I did... Attached Thumbnails

12th Apr 2020, 4:54 pm	#37
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status Unfortunately we don't know very much about the input and the trigger circuit of the 4094. Did you already check the boards under the CRT optically? Perhaps you find some strange looking things. You can also try to find supply voltage lines and measure them. Perhaps there are dead tantalum capacitors somewhere. You will have to do some reverse engineering. It's a pity we don't have a service manual for the 4094.. __________________ https://www.richis-lab.de

12th Apr 2020, 8:06 pm	#38
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status Sure I took a look at all, since I remove both top and bottom cover and unplugged boards to check all connections... without success. All seem OK. But again, trigger and auto-setup works well in normal operation, the issue is always with CAL involving trigger circuit. Even if 4074 and 4094 scope have not the same circuits, having info on 4074 trigger cal circuit may be helpful to understand how it works.

12th Apr 2020, 8:24 pm	#39
Noopy2014 Tetrode Join Date: May 2014 Location: Eichstätt, Germany Posts: 88	Re: Gould DSO 4094 SVG DEAD Status The 4074 and the 4094 are very different regarding the trigger. In the 4074 the trigger board has no calibration options. On the input amplifier board you can adjust the offset of the two signals going to the trigger board (CH1 and CH3). That's all. From the Input/Output board two analog signals are reaching the trigger board and control the trigger level. A serial bus controls the trigger board. A serial to parallel converter generates the signals that activate different couplings, low pass and high pass and TV functions. The only self calibration function that I can think of is a calibration of the trigger level. The trigger board is controlled by a MM5451. Perhaps the 4094 has also a MM5451. You can try to check whether that is still ok. => communication __________________ https://www.richis-lab.de

13th Apr 2020, 2:52 pm	#40
PBERTRAM Triode Join Date: Apr 2020 Location: Le Plessis-Robinson, Paris, France Posts: 17	Re: Gould DSO 4094 SVG DEAD Status Yes you're right (again), there is one MM5451 for the specific TV trigger circuit. But in fact there are many others on the board ! 4094 is just 4074 evolution and Gould certainly reused a lot from 4074 for the new one. But the more I have a look on the boards, the less I see burned device or whatever. Overall is rather clean with even very few suspect soldering.