Help fixing Iwatsu SS-5702 Analogue Osc dead channel

[ziplock9000]

I've had an old analogue 20Mhz, 2 channel Iwatsu SS-5702 for a few years and it does the job for what I need it for. The only problem is that channel 2 doesn't work.. not even in the slightest. There's no trace at all, no matter what you do with the interface.
I've checked the Ch1/Ch2 preamp board visually and everything looks to be ok. No exploded or burned out components, no broken wires or solder joints. Some of the pots are a little noisey and I'm in the process of cleaning them as part of a more general servicing.
I've had a look at the service manual, but I'm having trouble following it. The troubleshooting steps seem to be extremely basic and logic flow paths very limited.

I'd greatly appreciate any help or pointers from those of you with more experience.

Thanks.

[ziplock9000]

Well I did more visual inspection and cleaned all of the physical pots and dials with contact cleaner and channel 2 still didn't work.

So I began comparing the voltages from both channels when on the same settings with no input. I quickly hit a massive discrepancy between the two.
Pin 6 on the HA1127 on channel 1 (which works) read 0v, while P6 on the 2nd HA1127 is reading -11.7v!!!

This is when thing started to get weird. The official circuit diagram seems to be different to reality as far as how the HA1127s are connected for both channels.



This diagram represents channel 1 and 2, as they use the same diagram for both channels. As you can see R019 (ch1)/R119 (ch2) is supposed to be connected to pin 6 on the HA1127, when in fact it's connected to pin 2.

I've checked the orientation of the chip diagram to make sure it's not been flipped or revered.

This is confirmed because C016/C0116 connects to Pin 8.
Also in reality C017/C117 connects to R020/R120 not R024/R124

Anyway, I need sleep [/rant_over]

PS. Hmm, it might be because I have the service manual for the SS-5702A not SS-5702.. Which would make sense

[tautech]

Well I did more visual inspection and cleaned all of the physical pots and dials with contact cleaner and channel 2 still didn't work.

So I began comparing the voltages from both channels when on the same settings with no input. I quickly hit a massive discrepancy between the two.
Pin 6 on the HA1127 on channel 1 (which works) read 0v, while P6 on the 2nd HA1127 is reading -11.7v!!!

This is when thing started to get weird. The official circuit diagram seems to be different to reality as far as how the HA1127s are connected for both channels.



This diagram represents channel 1 and 2, as they use the same diagram for both channels. As you can see R019 (ch1)/R119 (ch2) is supposed to be connected to pin 6 on the HA1127, when in fact it's connected to pin 2.

I've checked the orientation of the chip diagram to make sure it's not been flipped or revered.

This is confirmed because C016/C0116 connects to Pin 8.
Also in reality C017/C117 connects to R020/R120 not R024/R124

Anyway, I need sleep [/rant_over]

PS. Hmm, it might be because I have the service manual for the SS-5702A not SS-5702.. Which would make sense

Let's try that without the img brackets:
http://imgur.com/a/21A6I

Can you give us a link to the schematic or a much larger snippet of it ?
Where do connections A and B go to ?

[ziplock9000]

Well I did more visual inspection and cleaned all of the physical pots and dials with contact cleaner and channel 2 still didn't work.

So I began comparing the voltages from both channels when on the same settings with no input. I quickly hit a massive discrepancy between the two.
Pin 6 on the HA1127 on channel 1 (which works) read 0v, while P6 on the 2nd HA1127 is reading -11.7v!!!

This is when thing started to get weird. The official circuit diagram seems to be different to reality as far as how the HA1127s are connected for both channels.



This diagram represents channel 1 and 2, as they use the same diagram for both channels. As you can see R019 (ch1)/R119 (ch2) is supposed to be connected to pin 6 on the HA1127, when in fact it's connected to pin 2.

I've checked the orientation of the chip diagram to make sure it's not been flipped or revered.

This is confirmed because C016/C0116 connects to Pin 8.
Also in reality C017/C117 connects to R020/R120 not R024/R124

Anyway, I need sleep [/rant_over]

PS. Hmm, it might be because I have the service manual for the SS-5702A not SS-5702.. Which would make sense

Let's try that without the img brackets:
http://imgur.com/a/21A6I

Can you give us a link to the schematic or a much larger snippet of it ?
Where do connections A and B go to ?

This is the service manual I'm using, although it's for the SS-5702A not SS-5702
http://www.radiopistoia.com/pdf/iwatsu.pdf

The schematic image I attempted to link was from page 160 in the pdf.

Thanks.

[ziplock9000]

I finally found the correct service manual and found the faulty component in 2 minutes.
A FET 2SK30A-Y is connected to the -12v rail on it's source and is outputting -11.7v on it's base and emitter. The one on the good channel is outputting almost zero. So it looks like it's shorted inside. Ordered 3 direct replacements from eBay for £2. Happy as a pig in S**t...

But looking ahead, that -11.7v goes into pin 3 (1G) of an NEC UPA61A.. Which if I'm reading the data sheet correctly V_gs(off), has a maximum negative value of -2.5v.. So it's almost certainly screwed. I'm hoping I've read the datasheet wrong.

Unfortunately this part just doesn't exist any more, there's no old ones I can find and there's no reference to any sort of replacement. So if it's dead I've just wasted a whole lot of time :/

[orbanp]

Hi ziplock,

I do not think that uPA61A went bad, the absolute max rating for the Vgs or Vgd is -40V!
The 2.5V is the Vgsoff spec, the voltage at which the current across the FET is zero.

Dual J-FETs are still available, just google them.
If push comes to shove, you can probably use two single J-FETS, just glue them together. It will work, with thermal matching specs a bit worse than a single chip dual FET.

Good luck, Peter

[ziplock9000]

Hi ziplock,

I do not think that uPA61A went bad, the absolute max rating for the Vgs or Vgd is -40V!
The 2.5V is the Vgsoff spec, the voltage at which the current across the FET is zero.

Dual J-FETs are still available, just google them.
If push comes to shove, you can probably use two single J-FETS, just glue them together. It will work, with thermal matching specs a bit worse than a single chip dual FET.

Good luck, Peter

Oh brilliant, Thanks for the clarification. I was just about to replace that part which was going to be expensive. I've replaced the bad transistor that now works. I've checked just about everything else out in the circuit and the only thing else giving wrong voltages is the HA1127. So I might just replace that and go from there.

[orbanp]

Hi ziplock,

The CA3045, CA3046 are equivalent to the HA1127.
There are also other equivalents to the CA3045-46.

Peter

[ziplock9000]

Hi ziplock,

The CA3045, CA3046 are equivalent to the HA1127.
There are also other equivalents to the CA3045-46.

Peter

Thanks mate.

[ziplock9000]

Hi ziplock,

I do not think that uPA61A went bad, the absolute max rating for the Vgs or Vgd is -40V!
The 2.5V is the Vgsoff spec, the voltage at which the current across the FET is zero.

Dual J-FETs are still available, just google them.
If push comes to shove, you can probably use two single J-FETS, just glue them together. It will work, with thermal matching specs a bit worse than a single chip dual FET.

Good luck, Peter

Well as a last leap of desperation, I unsoldered the good uPA61A and put it into channel 2 and the damn thing works. So my first hunch was right all along. The only problem is this part is impossible to find. Only one place sells it and it's expensive.

If you or anyone else could help my with an equivalent (I'd need pin assignments being a noob) it would be greatly appreciated.

[orbanp]

MODERATOR: Please move this thread over to the "Repair" section, the topic fits there a lot better!

Hi ziplock,

Sorry for the delay, I did not read this forum for a while!

That is a bummer that the dual FET is blown! Looks like whatever took out that Q001/101 FET, used just as a protection diode, that also damaged the dual FET!

When you say that uPA61A is expensive, just how expensive is that ? You should compare its price to the replacement value of a dual channel 20MHz analogue scope!

That aside, I did check a number of present day dual J-FETs. The only thing that those are not a good match for a uP61A. The uP61A has an Idss current about 2mA - 6mA, the present day dual FETs are in the 20mA to 30mA range! (They are also all SM devices!)
Idss is the current through a J-FET when the Vgs voltage is zero. That is a key parameter when replacing a J-FET. (See the J-FET on wikipedia: <https://en.wikipedia.org/wiki/JFET#/media/File:JFET_n-channel_en.svg>)

There are number of "older", single J-FETs available with such low range of Idss, through-hole, like the BF244A, BF245A, J113, 2N3822.

I would get a handful (10 - 20) J113 FETs (U$3.13 for 10 at DigiKey) and would select them for Idss. You would want to match two of them as close as possible. Connect the gate to the source, add 10VDC to the source and drain, and measure the current at the source. Pick two FETs where the current is the closest to each other and to the 2mA value!
The other beauty of the J113 is that it is a symmetrical device, drain and source are interchangeable, so when you glue them together at their flat faces you have a choice of pinouts! (use a drop of superglue)
But do measure the Idss as you would use them in the circuit!
Do not worry that you have no substrate (pins 4 and 8 on the uP61A) to tie to!

EDIT: Make that J-FET rather the BF244A or BF245A, the J113 is a switching FET! I will investigate that J113 further!

I will also plug that input circuit in to the simulator to see if there are any other issues!
I also do have an 5702 scope, for about 30 years now, running just fine.

Regards, Peter

[ziplock9000]

MODERATOR: Please move this thread over to the "Repair" section, the topic fits there a lot better!

Hi ziplock,

Sorry for the delay, I did not read this forum for a while!

That is a bummer that the dual FET is blown! Looks like whatever took out that Q001/101 FET, used just as a protection diode, that also damaged the dual FET!

When you say that uPA61A is expensive, just how expensive is that ? You should compare its price to the replacement value of a dual channel 20MHz analogue scope!

That aside, I did check a number of present day dual J-FETs. The only thing that those are not a good match for a uP61A. The uP61A has an Idss current about 2mA - 6mA, the present day dual FETs are in the 20mA to 30mA range! (They are also all SM devices!)
Idss is the current through a J-FET when the Vgs voltage is zero. That is a key parameter when replacing a J-FET. (See the J-FET on wikipedia: <https://en.wikipedia.org/wiki/JFET#/media/File:JFET_n-channel_en.svg>)

There are number of "older", single J-FETs available with such low range of Idss, through-hole, like the BF244A, BF245A, J113, 2N3822.

I would get a handful (10 - 20) J113 FETs (U$3.13 for 10 at DigiKey) and would select them for Idss. You would want to match two of them as close as possible. Connect the gate to the source, add 10VDC to the source and drain, and measure the current at the source. Pick two FETs where the current is the closest to each other and to the 2mA value!
The other beauty of the J113 is that it is a symmetrical device, drain and source are interchangeable, so when you glue them together at their flat faces you have a choice of pinouts! (use a drop of superglue)
But do measure the Idss as you would use them in the circuit!
Do not worry that you have no substrate (pins 4 and 8 on the uP61A) to tie to!

EDIT: Make that J-FET rather the BF244A or BF245A, the J113 is a switching FET! I will investigate that J113 further!

I will also plug that input circuit in to the simulator to see if there are any other issues!
I also do have an 5702 scope, for about 30 years now, running just fine.

Regards, Peter

When I say expensive, I mean around $30 USD from the one and only company that seems to sell it. Even then, the company looks dodgy as it seems to have a large supply of a now defunct part that no other supplier or individual on the planet seems to have. The scope itself is not worth much more.

Thank you for investigating this. As a beginner I don't know how to find a compatible replacement. I'll order some BF244A or BF245A's and try and follow your testing recommendation.

Thanks.

[orbanp]

Hi ziplock,

Yes, I would stay with the BF244A, BF245A FETs. The J113 are mostly characterized in the data sheet for switching applications.
Note that the BF244A, BF245A FETs are not symmetrical, so watch out for source and drain pins when two are attached.
With the R013/113 pot you can set the DC balance. To do that set the vertical position pot on the front panel to the middle, then move the beam to the middle of the screen with that trim pot.
If there is not enough adjustment range with the trimpot then you can try swapping around the FET pair, what was at the bottom would go to the top and vice-versa.
You could also change R012/112 if you run out of adjustment range.
But I would try to get the best Idss match.

Good luck, Peter

[ziplock9000]

I've just ordered 10 BF245A from eBay and should be on the slow boat from China.
Here's my plan of attack, but unfortunately the pins cross over onto the board, so I'll have to rig something up.