On SD's the cnroller chip constantly moves where you are writing to manage the wear on the drive. A bigger drive will last much longer. The numbers I have seen suggest a 128 GB or larger drive will not wear out in normal use in less than 10 years. Other stuff suggest they are as reliable as spinning disks. So great but not a substitute for backups. I would probably buy a new Samsung SSD. I have some lower grade SSD's that don't work well in a laptop. And two 250 GB SSD's that work great, both Samsungs. The time you waste troubleshooting that stuff is never worth it.
A1 Board Finishing
Trying to finish up the A1 board.
Popped out the A1U3 buffer amp,
installed socket, then opa134.
Included with the change:
C49 - changed to 0.1 uf PPcap
C15 - removed
C17 - removed
C16 - removed
C14 - removed
Not much difference but higher 2H.
Then removed and installed LT1468 with
pin 8 removed. Not good. harmonics everywhere.
Funny though, nothing shows up on the scope,
only on QA400.
Thinking I need to put C15, 17, 16, 14 back in?
For either opa134 or LT1468?
Does the LME49710 metal can work here? for the Buffer amp?
The buffer amp is controlled input from R3 (the Osc level pot)
don't know if small 10 ohm from pin 2, to pin3 would help
on the opa134.
Also changed C40, C41s to the 100uf 50V bi polar Nichicon.
OR
Is compensation still needed?
Trying to finish up the A1 board.
Popped out the A1U3 buffer amp,
installed socket, then opa134.
Included with the change:
C49 - changed to 0.1 uf PPcap
C15 - removed
C17 - removed
C16 - removed
C14 - removed
Not much difference but higher 2H.
Then removed and installed LT1468 with
pin 8 removed. Not good. harmonics everywhere.
Funny though, nothing shows up on the scope,
only on QA400.
Thinking I need to put C15, 17, 16, 14 back in?
For either opa134 or LT1468?
Does the LME49710 metal can work here? for the Buffer amp?
The buffer amp is controlled input from R3 (the Osc level pot)
don't know if small 10 ohm from pin 2, to pin3 would help
on the opa134.
Also changed C40, C41s to the 100uf 50V bi polar Nichicon.
OR
Is compensation still needed?
Save the metal cans for audio. The plastic dips are just fine for this application. Low level harmonics are a sign something is unhappy. Probably a low level oscillation.
I spent the weekend doing something similar with a Boonton 1121. It can be very frustrating when it doesn't work as expected. The LME497XX parts seem to be pretty tolerant and fit most everywhere. Be careful when replacing a FET opamp. The bias current is much higher on the bipolar and will cause unexpected results on high impedance circuits.
I spent the weekend doing something similar with a Boonton 1121. It can be very frustrating when it doesn't work as expected. The LME497XX parts seem to be pretty tolerant and fit most everywhere. Be careful when replacing a FET opamp. The bias current is much higher on the bipolar and will cause unexpected results on high impedance circuits.
The compensation parts in the feedback loops of the HA2625s (? right part?) are not needed by the 1468 or OPA134. A small cap, say 15pF, in parallel with the feedback resistor that sets the gain will be helpful with the 1468. I honestly don't remember the buffer amp circuit, but an LME part should work well there, since, IIRC, it is driven from a relatively low-Z source
I would be tempted to try an LME49990 here. But only if I had one on hand.
Putting the lids back on make a difference with the oscillator.
Putting the lids back on make a difference with the oscillator.
Where I'm At...
Here is the latest:
1. From the scan of the schematic, buffer amp
2. Here is the board.
3. Here's the parts layout.
Here is the latest:
1. From the scan of the schematic, buffer amp
2. Here is the board.
3. Here's the parts layout.
Last edited:
I just removed the caps that feedback to the original amp and left the rest as-is for the 1468. There was a lot of phase shaping to extend the operation to 100KHz. I didnt care about that and tried to get lowest thd up to 10-20KHz only. You can extend the BW but the thd goes up... limit the BW and thd (+N) goes down. There was not so much to be gained with the buffer.... not my first order choice of mods.
THx-RNMarsh
THx-RNMarsh
"Who makes a chassis punch with the little hold down, anti rotation notch?"
Just to make your search easier, the anti rotation notch is said to be 'keyed'.
Try searching 'keyed punch'.
Lots of Greenlee here.
http://ca.mouser.com/Tools-Supplies/Tools/Punches-Dies/_/N-5gge?P=1yyq1d7
Just to make your search easier, the anti rotation notch is said to be 'keyed'.
Try searching 'keyed punch'.
Lots of Greenlee here.
http://ca.mouser.com/Tools-Supplies/Tools/Punches-Dies/_/N-5gge?P=1yyq1d7
Last edited:
@Richard, I re-read that, not lots to be gained, with that A1U3,
sometimes iffy. I'm working on a board area at a time, which
might not be the best way to go about it.
As far as A1u3 buffer amp, maybe .1uf is too large for feedback,
maybe go back to stock 15 pf on C49.
I guess I'm taking out too many components, I should probably
leave more in. Live and learn.
There is a lot to digest in those 2000 plus posts relating to HP339a mods
and there is some going back and forth to what is done, undone,
wrong board/nomenclature etc, etc, it is what it is. I'm thinking this
might help some other DIYers going through it. Condensing it and
clarifying it for those that follow.
I'll see how the cap change back works.
So, after this cap change, what would be the next priorities?
The TLO74 for the filters A2U1, and A2 U10 LT1468 with 100kHz
bandwidth mod on the feedback circuit.
Then A2U4 with another LT1468,
with C16/C17 coupling caps replaced?
Wondering about the Auto Set Level opamp series?
will the LME49740?
@DAVADA
sometimes iffy. I'm working on a board area at a time, which
might not be the best way to go about it.
As far as A1u3 buffer amp, maybe .1uf is too large for feedback,
maybe go back to stock 15 pf on C49.
I guess I'm taking out too many components, I should probably
leave more in. Live and learn.
There is a lot to digest in those 2000 plus posts relating to HP339a mods
and there is some going back and forth to what is done, undone,
wrong board/nomenclature etc, etc, it is what it is. I'm thinking this
might help some other DIYers going through it. Condensing it and
clarifying it for those that follow.
I'll see how the cap change back works.
So, after this cap change, what would be the next priorities?
The TLO74 for the filters A2U1, and A2 U10 LT1468 with 100kHz
bandwidth mod on the feedback circuit.
Then A2U4 with another LT1468,
with C16/C17 coupling caps replaced?
Wondering about the Auto Set Level opamp series?
will the LME49740?
@DAVADA
For the buffer I think the 4.7K resistor is an effort to match impedances between the + and the - input. For the LME49710 I'm not sure that will help and it will add noise. At least remove the .1 uF cap since it will cause issues with an impedance variation in the middle of the operating band. I would just jumper it. Try it both jumpered and 4.7K only. If you are using an AD797 or LME49990 you would need a 100 Ohm resistor for stability. I think the original circuit was to deal with either some frequency extension or marginal stability problem. The modern opamps will not have bandwidth problems. There may be issues from the capacitive loading on the 10K pot (that's where the 5K resistor value came from). I would rather see a 1K pot (and 257 Ohm resistor) if the previous stage can drive it. The high impedance stuff is just an opening for noise and response variations.
Thanks Ramya & Gents.
That little key nibbler is only 475USD...
...or is that Canadian Dollar which costs more
when converting.
That little key nibbler is only 475USD...
...or is that Canadian Dollar which costs more
when converting.
Last edited:
/@Demian, for the Buffer AMp, I currently have the opa134 installed.
The LT1468 didn't work so well.
But, on the feedback I'll give it a shot of 4.7K and maybe the 15 pf cap.
for the opa134 in the buffer amp.
Next question, as this gives me the most problems I don't know how
to recognize the typ of drive and loads these op amps will be seeing.
I don't know what to look at in the schematic nor the layout.
Even looking at the schematics, go figure.
So now that we are talking changes that I understand finally, I hope,
What compensation would optimize the performance of the quad LME497740 in the Summing Amp, Integrator, Difference Amp.
Guidancd and Assistance
Night All, have a great week.
@Richard when will lyou have some time at the condo open for visitors?
The LT1468 didn't work so well.
But, on the feedback I'll give it a shot of 4.7K and maybe the 15 pf cap.
for the opa134 in the buffer amp.
Next question, as this gives me the most problems I don't know how
to recognize the typ of drive and loads these op amps will be seeing.
I don't know what to look at in the schematic nor the layout.
Even looking at the schematics, go figure.
So now that we are talking changes that I understand finally, I hope,
What compensation would optimize the performance of the quad LME497740 in the Summing Amp, Integrator, Difference Amp.
Guidancd and Assistance
Night All, have a great week.
@Richard when will lyou have some time at the condo open for visitors?
Open for businesss right now 🙂
BTW -- near the end of those 2000 lines of work.... we did a summery of what to do first for most bang for the buck (and time involved). Might be best to start with those suggested changes only... and afterwards try the 2ndary issues and affects.
THx-RNMarsh
BTW -- near the end of those 2000 lines of work.... we did a summery of what to do first for most bang for the buck (and time involved). Might be best to start with those suggested changes only... and afterwards try the 2ndary issues and affects.
THx-RNMarsh
I would do a load test on the oscillator op amp. See how low you can go. Then size the pot from there. I found the pot to be a significant addition to the distortion contribution. Not necessarily from loading.
Parasitic osc will raise the distortion. At least the measurement distortion. It often looks on fft like the op amp is clipping. That kind of display. The noise in the spectrum will increase. It generates a heavy IMD. The QA bandwidth in not wide enough to see the parasitic. You just see the products. What you see as a rise of THD might be an op amp on the edge of osc.
Viewed on a scope parasitic osc might be small and looks like the trace is fat like high frequency noise. You can use an RC high pass filter to filter the fundamental out making the parasitic osc more visible on a scope.
Parasitic osc will raise the distortion. At least the measurement distortion. It often looks on fft like the op amp is clipping. That kind of display. The noise in the spectrum will increase. It generates a heavy IMD. The QA bandwidth in not wide enough to see the parasitic. You just see the products. What you see as a rise of THD might be an op amp on the edge of osc.
Viewed on a scope parasitic osc might be small and looks like the trace is fat like high frequency noise. You can use an RC high pass filter to filter the fundamental out making the parasitic osc more visible on a scope.
Try just drilling a hole for the shaft and another small hole for the key.
If it is wobbly then use a drop of super glue
If needed later the super glue can be dissolved with acetone.
I looked and couldn't find the order of mods/fixes.
I did find an order frm posts: 1391 - 1404.
Those are for the A1 board and the A4 board.
Then at post #2493 for the Notch Amp A3 U3 opa1461.
Wondering if I can get away with the opa134 there.
either that or work with the little breakout etc,
the way David suggested.
myhrrhleine, thanks for the suggestion. Usually when I try
that stuff I get a wobble then a scrape across what ever I
don't want to scratch. Tape can be your friend I've found.
I did find an order frm posts: 1391 - 1404.
Those are for the A1 board and the A4 board.
Then at post #2493 for the Notch Amp A3 U3 opa1461.
Wondering if I can get away with the opa134 there.
either that or work with the little breakout etc,
the way David suggested.
myhrrhleine, thanks for the suggestion. Usually when I try
that stuff I get a wobble then a scrape across what ever I
don't want to scratch. Tape can be your friend I've found.
Priorities
@Richard
Still floating about looking for them.
Correct me if I'm wrong, post 2000 +-400
and still missing it.
I must be blind. Somewhere in the post # 950 - #1404
There seems to be a lot of meat that DIYer can use.
After I finish the basic stuff on A4 board.
@ Demian, I'll make the changes to the A1U3 4.7K.
The opa134 data sheet has + - input at 10 ohms pin2, 3.
4.7K is aways away and it is discussed in data sheet to not
make that resistor too high.
Yes, I'm actually reading and studying them to figure out
what is going on.
I'm still having a problem trying to identify stage impedances
and when to use/not use FETs or bipolar opamps.
I now know there is the trade off between voltage noise and
current noise.
@Davada,
For the LT1468 and setting feedback per the datasheet, or was
that Demian? From what is stated there, at least for the 100 kHz
is a 1000 pf & 22.1K resistor But, it its output is loaded with a 1M ohm
resistor. So, I'd set this up as a band pass filter?
@Richard
Still floating about looking for them.
Correct me if I'm wrong, post 2000 +-400
and still missing it.
I must be blind. Somewhere in the post # 950 - #1404
There seems to be a lot of meat that DIYer can use.
After I finish the basic stuff on A4 board.
@ Demian, I'll make the changes to the A1U3 4.7K.
The opa134 data sheet has + - input at 10 ohms pin2, 3.
4.7K is aways away and it is discussed in data sheet to not
make that resistor too high.
Yes, I'm actually reading and studying them to figure out
what is going on.
I'm still having a problem trying to identify stage impedances
and when to use/not use FETs or bipolar opamps.
I now know there is the trade off between voltage noise and
current noise.
@Davada,
For the LT1468 and setting feedback per the datasheet, or was
that Demian? From what is stated there, at least for the 100 kHz
is a 1000 pf & 22.1K resistor But, it its output is loaded with a 1M ohm
resistor. So, I'd set this up as a band pass filter?