Re: Power supply problem with Tek 2220
I saw your 2220 question the first time you posted it a few days ago, and I
did not have a nice short answer for you. Since you have not gotten the
answer
you were looking for, let me comment a bit on this.
First off, I don't own a 2220 or have the service manual for it, and none of
the download sites I checked seem to have it either, including
ArtekMedia.com having it for purchase. So I will assume it is similar to the
better 22xx's such as the 2235, at least around the power supply. I will
also assume that you do have a service manual to work with. U930 monitors
the voltage drop across R907, and in your case it probably thinks the
current is too high after a few minutes, so it shuts down.
I have worked on lots of 22xx scopes and I have not seen the value of R907
drift and cause this problem, but others have reported that as a
possibility. Odds are, that this is not your problem.
An over current situation can be caused by anything after R907 within the
inverter, the transformer itself or on the secondary side of the main
transformer.
How you go about finding the problem depends on what equipment you have
available, and what your testing preferences might be.
If you don't have much to work with, just start changing components. In my
experience the components in the right rear corner of the mainboard cause
the most grief so I would start there with C925, C942 and C943. If you have
some decent soldering equipment you can change all three capacitors quickly
without removing the mainboard.
If you have access to an ESR meter you could check all of the aluminum
electrolytic capacitors in circuit. Most of the electrolytics give no
trouble, especially the six(?) 840 uF caps in front of the heat sink for
Q9070 (but you would want to check all of the electrolytic caps if you had
an ESR meter handy).
Another approach would be to put in the 43 volts DC from an external power
source until something got hot or went up in smoke. If your power supply
allowed you to adjust the current limit you could sneak up on it and just
get the problem part(s) nice and warm. You will need to at least disconnect
Q9070 when you apply the external 43 VDC at TP940 and TP 950 from the bottom
or top of the mainboard. I lightly tack on two wires to the bottom side to
make the 43 VDC connections. You can use the scope in its normal way when it
is powered with the external 43 VDC.
The external 43 VDC idea came from a fine Tekscopes member named Hakan, many
years ago, and he has an excellent document you should read at:
A couple of years ago I made up a document listing some commonly available
components you might use to replace some of the original Tektronix part
numbers that often fail. I can send you that directly or you can find it in
the Tekscopes Message archive if you like.
Another thing you might do is take resistance readings from chassis ground
to each of the voltage test points before you get very far into this repair
(with the scope not connected to the mains of course!).
There are lots of other ideas to add to this, but this should get you
started.
This will be a simple fix, so just take your time and understand what you
are doing.
tom jobe...
toggle quoted message
Show quoted text
----- Original Message ----- From: "Tan Chor Ming" <jonray03@...> To: <TekScopes@...> Sent: Friday, January 20, 2012 6:45 AM Subject: Re: [TekScopes] Power supply problem with Tek 2220 Has anyone seen this behaviour repairing Tek2000 series scope power supply?
Behaviour
1. When power on from cold, it will fire up properly and stay on for a few
minutes
2. Once it warms up, the power supply shut down and try to on again and
it continues in this on/off mode
3. Pre-regulator out = 43V (between TP940 and TP950)
4. All voltages when the unit is stable for a few minutes are +8.6, -8.6,
+5.2, +100 are all ok.
What are likely problems before I start ripping the components from the
circuit?
ttesenq@... suggested that it could be the pre-regulator MOSFET
P9070.
I though that it could be a electrolytic cap but it doe not look likely
because electrolytic failure would not have allowed the unit to fire up for a few minutes
It is more likely a active component that weaken with increase temperature, so ttesenq@... could be right.
This would mean Q947 and Q946 are also likely components
Any other possible lead of similar experience.
|
Re: TR502 7L13/14 etc Lemo connector
Hello, --- In TekScopes@..., "Peter" <pdusel@...> wrote: Would you care to share the distributor you used, I went to lemo.com and selected the nearest office contact details and sent them an email inquiry. and I'm almost afraid to ask, the price? I am shamed to admit that "right now on my desk" actually means they have been buried under other parts for about a year now. Since you reminded me about the connectors I even went to local shop today to buy some 6 wire shielded cable. Sort of progress in other words. About the price, I have "archived" the invoice somewhere but if I run into it I'll add the cost to this thread. If I remember correctly the total was somewhere between 150 and 200 eur for six connectors and their bend reliefs without VAT. That FGG.2B was the most expensive of them at 40 (maybe more) eur IIRC. Bend reliefs were in order of eur or two each. That leaves at least 25 eur each for the 6-pin connectors. I hope this helps, --- In TekScopes@..., "nukescope" <vtp@> wrote:
Hello,
--- In TekScopes@..., "Peter" <pdusel@> wrote:
Does anybody know what the part number is for the six pin Lemo connector used on the logic cable used to connect a TR-502 to a 7L13/14?
I happen to have a bag of those right on my desk and reading from the bags I have
FFA.1S.306.CLAC42Z - 6-pin connector
GMA.1B.040.DG - bend relief
Sorry, no excess - I bought four of those for two cables. Had to order them from local Lemo office.
...
Since the other typical application for LEMO connectors comes up sometimes I am taking the opportunity to include the same information for CG511/CG5001 output cable.
I am reading FGG.2B.802.CLAM31Z for connector and GMA.0B.025.DG for bend relief for RG174/316. Again, these are for CG5001 calibration generator.
|
Hi to All,
Tek used the 8393 Nuvistors in the Type 1A6 and 1A7. Look at the Circuit Description of both Plug-Ins and you know why. The so called differential adjustment of heater voltages. In addition lower heater current is lower hum/noise (the reason Tek used 8416's in the 1S1).
Egge Siert
|
Re: 7T11 Sequential/Random Problem
Probably meaning it's only there at the fast ranges, 500 ns and 50 ns?
The Rate meter can cope with an enormous range of repetition rates, split in 3 parts IIRC (more or less logarithmic). Because of that it's very well possible that the effect is confined to the faster timing ranges. Apart from a dozen other possible reasons of course.
Albert
toggle quoted message
Show quoted text
The effect was not visible for risetimes longer than about 500 ps.
|
Re: SG503 - Troubleshooting
Hi Patrick,
I have some good news: my SG503 is fixed and running. With your and David's help, I identified Q180 (2N3906) as having failed . C180 was also damaged (open). R180 and R216 had signs of cracked epoxy but their values were good. I replaced all of these components and the unit is running fine.
The scenario that seems plausible to me is that the original culprit was Q180, failing intermittently and causing current surges. These surges blew C172 (1mF tantalum) a couple of times. C180 was good initially because I had replaced with a new 1mF/50V tantalum along with C172. Eventually when I applied external power, Q180 failed completely taking C180 along with it.
For future reference, here are some measurements for Q180 and Q190.
Bad Q180 Good Q180
V(B-E)=0.288 V, R(B-E)=350 Ohms V(B-E)=1.481 V, R(B-E)=6.0 KOhms
V(E-B)=0.273 V, V(B-E)=0.673 V
V(B-C)=0.289 V, R(B-C)=352 Ohms V(B-C)=0.750 V, R(B-C)=4.92 KOhms
V(C-B)=0.275 V, V(C-E)=0.643 V
V(C-E)=0.000 V, R(C-E)=6.5 Ohms V(C-E)=0.666 V, R(C-E)=1.63 KOhms
V(E-C)=0.000 V, V(E-C)=1.305 V
Q190
V(B-E1)=0.701 V, V(E1-B)=1.339 V
V(B-E2)=0.735 V, V(E2-B)=1.340 V
V(C-E1)=0.719 V, V(E1-C)=0.609 V
V(B-C)=0.694 V, V(C-B)=1.087 V
V(C-E2)=0.748 V, V(E2-C)=0.614 V
Thanks again,
-Achilles
toggle quoted message
Show quoted text
--- In TekScopes@..., "Patrick Wong" <patwong3@...> wrote:
Hi Achilles,
I'm wondering whether you've made further progress on your SG-503? Would like to hear the outcome when you have a chance.
Thanks,
Patrick Wong AK6C
|
Re: 7T11 Sequential/Random Problem
Is this the kind of application where the differences between
saturating (TTL) and non-saturating (ECL) logic would apply? I wonder
if the extra jitter is from poor power supply rejection.
toggle quoted message
Show quoted text
On Fri, 20 Jan 2012 19:57:34 -0000, "Mike" <zuckerme@...> wrote: BTW a 74LS14 with a resistor and mica capacitor provide a general-purpose 100 ns pretrigger delay. It seems stable to better than 20 ps rms if you happen upon a "good" inverter.
|
Re: 7T11 Sequential/Random Problem
I recall the behavior's independent of rep rate (at least between 1 kHz or so and 1 MHz). I initially saw it with my homebrew avalanchers, but it also was apparent with 067-0681-01 (driven at whatever rate you please), HP213B, 284, and S-52. I never tested it on my 109 because life is too short.
The effect was not visible for risetimes longer than about 500 ps.
BTW a 74LS14 with a resistor and mica capacitor provide a general-purpose 100 ns pretrigger delay. It seems stable to better than 20 ps rms if you happen upon a "good" inverter.
I also have a 7M11 delay line, but it's sluggish. I think the cable shield is corroded.
All in all I would love to fix my 7T11 and 7T11A but not clear where to start.
toggle quoted message
Show quoted text
--- In TekScopes@..., "Albert" <aodiversen@...> wrote:
Hi Mike,
Strange things in the dot pattern happen indeed in Random mode, but I don't remember a case of time-reversal.
I think much depends on the repetition rate of your signal generator. I think the 264 and S-52 gave me no problems. I don't have faster repetition fast pulses.
Long ago I have seen negative rise and fall time with 3S76/3T77A. And that was sequential mode while showing the square wave output of a 284.
Albert
--- In TekScopes@..., "Mike" <zuckerme@> wrote:
David, Albert-
I have vaguely similar issues with both my 7T11 and 7T11A in random mode. I see large gaps in the horizontal dot density, and even misplaced (time-reversed!!) dots. For example by moving 7S11 delay, 7T11 time offset, and trigger position I can display a 100 ps step with positive, zero or even negative risetime (all with clean repeatable triggering).
---
Mike
|
r. Unfortunately Q203 is one of the more tedious-to-replace transistors in the HP130C, so at present it's put-off-today-what-you-can-do-later :), with the added excuse: experimentally lets see how long such a simple emergency fix is good for. Berj / KI3U
To: TekScopes@... From: gumbear@... Date: Fri, 20 Jan 2012 18:29:40 +0000 Subject: [TekScopes] Re: Nuvistors.
?
Q203 may have developed more leakage over the years and will eventually have to be replaced. I'd look for a replacment just to be ready for when it does.
Arden
> By way of a little OT, mine after decades of routine use last year developed, upon warm-up, folding and contracting of the trace. The problem turned out to be transistor Q203 in the horizontal amplifier. The extend-its-life-fix was to put a 10-fins light sheet aluminum heatsink on Q203.
|
Re: 7T11 Sequential/Random Problem
Hi Mike,
Strange things in the dot pattern happen indeed in Random mode, but I don't remember a case of time-reversal.
I think much depends on the repetition rate of your signal generator. I think the 264 and S-52 gave me no problems. I don't have faster repetition fast pulses.
Long ago I have seen negative rise and fall time with 3S76/3T77A. And that was sequential mode while showing the square wave output of a 284.
Albert
toggle quoted message
Show quoted text
--- In TekScopes@..., "Mike" <zuckerme@...> wrote:
David, Albert-
I have vaguely similar issues with both my 7T11 and 7T11A in random mode. I see large gaps in the horizontal dot density, and even misplaced (time-reversed!!) dots. For example by moving 7S11 delay, 7T11 time offset, and trigger position I can display a 100 ps step with positive, zero or even negative risetime (all with clean repeatable triggering).
---
Mike
|
Q203 may have developed more leakage over the years and will eventually have to be replaced. I'd look for a replacment just to be ready for when it does.
Arden
toggle quoted message
Show quoted text
By way of a little OT, mine after decades of routine use last year developed, upon warm-up, folding and contracting of the trace. The problem turned out to be transistor Q203 in the horizontal amplifier. The extend-its-life-fix was to put a 10-fins light sheet aluminum heatsink on Q203.
|
Re: 7T11 Sequential/Random Problem
Hi David,
It's not an elongated dot. It's simply a small segment of the sine waveform. If you scroll through the scan positions, you will see that "dot" follows the shape of the sine wave.
Therefore maybe jitter is a nasty term (it does not degrade the REP scan). Specific causes are repetition period jitter (hum) in your signal generator, and jitter in the delay times produced by the Rate meter. Both are "corrected" to show the dot at the proper horizontal position, i.e. the dot position is determined according to the recorded time difference between strobe and (nearby) trigger event.
Albert
toggle quoted message
Show quoted text
.... and I've now got a fairly stable trace but the dot is elongated to about 0.4 vertical divisions at the zero crossing points when in manual scan mode, is this normal or do I still have an excessive amount of trigger jitter?
---
David Partridge
|
Re: OT XOR doesn't work as it should
I have seen them used though where profile was not a consideration. Tektronix of course early on had their own version with those low profile Teflon(?) insulated collets before they switched to flip-chips or whatever that other packaging style is. I have not needed to take one of those apart yet. I am of the same mind as the other commenter: if the prototype works with sockets then the production model will work even better and have more margin without if necessary. On Fri, 20 Jan 2012 06:53:11 -0000, "phosphorphile" <gumbear@...> wrote: Yes, but I think that technology was developed to reduce part heights for denser board stacking, not for faster fasts.
Arden
They make low profile sockets which work as you describe. The
machined collets have no pin on the bottom and sink directly into
larger than normal holes in the board.
|
Hi?
Allow me to offer a different opinion on this isolation matter. I worked on a 7D13 with an isolated(floating) section this weeikend. I used my isolation gear so that I could tie my scope ground(common) to the floating common. In such cases isolation is desirable. In the 7603, the power supply transformer provides all the isolation you need. Using another level of isolation allows the 7603's chassis to float and that is not good. It is an unneeded safety hazard.
Jerry Massengale
jmassen418a@...
toggle quoted message
Show quoted text
-----Original Message-----
From: David Miles
To: TekScopes
Sent: Fri, Jan 20, 2012 5:51 am
Subject: [TekScopes] Re: 7603 no trace
?
Hello Magnus.
I do have a piece of equipment like that , this is what it looks like
Can you please explain some more. How can I find out if it is isolating,
and how/why should I be using it?
Thanks.
David.
|
Re: TR502 7L13/14 etc Lemo connector
Hi!
Thanks much for the information! It's in the archive now, so the next to search on this will find it.
Would you care to share the distributor you used, and I'm almost afraid to ask, the price?
Thanks again,
Pete
toggle quoted message
Show quoted text
--- In TekScopes@..., "nukescope" <vtp@...> wrote:
Hello,
--- In TekScopes@..., "Peter" <pdusel@> wrote:
Does anybody know what the part number is for the six pin Lemo connector used on the logic cable used to connect a TR-502 to a 7L13/14?
I happen to have a bag of those right on my desk and reading from the bags I have
FFA.1S.306.CLAC42Z - 6-pin connector
GMA.1B.040.DG - bend relief
Sorry, no excess - I bought four of those for two cables. Had to order them from local Lemo office.
...
Since the other typical application for LEMO connectors comes up sometimes I am taking the opportunity to include the same information for CG511/CG5001 output cable.
I am reading FGG.2B.802.CLAM31Z for connector and GMA.0B.025.DG for bend relief for RG174/316. Again, these are for CG5001 calibration generator.
|
Re: Power supply problem with Tek 2220
At least check the incircuit power supply electrolytic capacitors for ESR. If you don't have an ESR tester, buy one or build one. There is lots of ESR information on the web and also in the archives of this yahoo group. ?You definitely need one if you're considering working on this type of electronic equipment and problems. Otherwise give the problem to someone else who has this essential test equipment and basic electronic knowledge to assist in diagnosing your problem. You are over your head without basic skills and test equipment.
From: Tan Chor Ming
To: TekScopes@...
Sent: Friday, January 20, 2012 8:45 AM
Subject: Re: [TekScopes] Power supply problem with Tek 2220
?
Has anyone seen this behaviour repairing Tek2000 series
scope power supply?
?
Behaviour
1. When power on from cold, it will fire up properly and
stay on for a few minutes
2.? Once it warms up, the power supply shut down and
try to on again and it continues in this on/off mode
3.? Pre-regulator out = 43V? (between TP940 and
TP950)
4.? All voltages when the unit is stable for a few
minutes are +8.6, -8.6, +5.2, +100 are all ok.
?
What are likely problems before I start ripping the
components from the circuit?
?
?ttesenq@...
suggested that it could be the pre-regulator MOSFET P9070.
?
I though that it could be a electrolytic cap but it doe not
look likely because electrolytic failure would not have allowed the unit to
fire up for a few minutes
It is more likely a active component that weaken with
increase temperature, so ttesenq@...
could be right.?
This would mean Q947 and Q946 are also likely components
?
Any other possible lead of similar experience.
|
Hello Magnus.
I do have a piece of equipment like that , this is what it looks like
Can you please explain some more. How can I find out if it is isolating, and how/why should I be using it?
Thanks.
David.
|
Don Lewis wrote Fri, 20 Jan 2012 04:36:14 -0800 :
?
" My old HP-130C scope?has two 7586 Nuvistors
in it, btw. "
A really fine rig. I read somewhere the HP-130C is regarded by some as the last good oscilloscope HP made.
By way of a little OT, mine after decades of routine use last year developed, upon warm-up, folding and contracting of the trace. The problem turned out to be transistor Q203 in the horizontal amplifier. The extend-its-life-fix was to put a 10-fins light sheet aluminum heatsink on Q203. I coated a light film of Desitin on the mating surfaces. The scope has been back to normal ever since. Desitin, which I read about somewhere on the net as a handy heatsink compound, is the baby rash creme found in supermarkets etc., it is much cheaper than the ZnO formulations marketed specifically as heatsink compound. Subjectively, I've had better luck with Desitin than the latter, perhaps because the Desitin is creamier and therefore easier to get on there as a thin film.
Berj / KI3U
|
Re: 7T11 Sequential/Random Problem
Hi Albert,
Pointing me to triggering was an excellent judgement call.
I fiddled a bit with the bias controls on the various TDs and managed to get a stable display at some trigger levels, but it was very "twitchy". Using the classic finger test, it looked awfully like the arming TD was oscillating.
I shortened the top lead to the diode and readjusted the arming bias, and I've now got a fairly stable trace but the dot is elongated to about 0.4 vertical divisions at the zero crossing points when in manual scan mode, is this normal or do I still have an excessive amount of trigger jitter?
When I replaced the arming and trigger diodes, the tinning on the square pad to which they were mounted came off with the solder and I had stick a small square of copper foil onto that to allow me to solder them in place :( Maybe that's added some undesirable stray capacitance to the circuit.
I still can't get rid of the left shift of the trace on the equivalent time ranges when I switch from sequential to random (about 2.5 divisions) or as I turn the time/div control clockwise within a sweep range - clearly I have one of the calibration adjustments incorrect (but which), or there's another problem.
I've added a couple of pictures to the album showing the current state of the 20MHz signal in random mode (including left shift) and of the slightly elongated dot at the zero crossing in manual scan mode.
Regards,
David Partridge
toggle quoted message
Show quoted text
-----Original Message----- From: TekScopes@... [mailto:TekScopes@...] On Behalf Of Albert Sent: 20 January 2012 13:39 To: TekScopes@... Subject: [TekScopes] Re: 7T11 Sequential/Random Problem Hi David, The RHS is also far from good. At those 20 MHz you should see no quality difference between sequential and random. I can more or less simulate your random waveform by playing with the trigger Level (also with LHS dim and/or reduced size, RHS noisy but more normal). So I guess it's a triggering problem. Random is much more sensitive to missed or wrong time trigger events. I used 20MHz from SG503, internal triggering, range 500 ns, speed 50 ns/div. Albert Hi folks,
Please could you take a look at the pictures in here:
<
st>
All working fine in sequential mode, but in random mode only the RHS of the display is correct?
Does anyone have any ideas what could be causing this?
Thanks
Dave
------------------------------------ Yahoo! Groups Links
|
Re: 7T11 Sequential/Random Problem
David, Albert-
I have vaguely similar issues with both my 7T11 and 7T11A in random mode. I see large gaps in the horizontal dot density, and even misplaced (time-reversed!!) dots. For example by moving 7S11 delay, 7T11 time offset, and trigger position I can display a 100 ps step with positive, zero or even negative risetime (all with clean repeatable triggering).
Since the incoming trigger is shared with sequential mode, I think I must be noise or crosstalk from the sample clock getting into the delta time converter or the random sampling strobe. Perhaps some bad decoupling caps?
This problem renders random mode all but useless for risetime measurements, but I haven't had time (or enough extenders) to dig into it. So far I've just gotten by with sequential mode which is fine.
If you have a fast pulse source, I'd be interested to see if you can check how random mode works with that. With CW sines it's hard to interpret where the trigger is occurring (they all look alike).
I won't have access to my rig until next week but can put up some pictures then if it helps.
Mike
toggle quoted message
Show quoted text
--- In TekScopes@..., "Albert" <aodiversen@...> wrote:
Hi David,
The RHS is also far from good. At those 20 MHz you should see no quality difference between sequential and random. I can more or less simulate your random waveform by playing with the trigger Level (also with LHS dim and/or reduced size, RHS noisy but more normal). So I guess it's a triggering problem. Random is much more sensitive to missed or wrong time trigger events. I used 20MHz from SG503, internal triggering, range 500 ns, speed 50 ns/div.
Albert
Hi folks,
Please could you take a look at the pictures in here:
<>
All working fine in sequential mode, but in random mode only the RHS of the display is correct?
Does anyone have any ideas what could be causing this?
Thanks
Dave
|
Good Day David,
to amend: You may also want to get yourself an isolation transformer or an *isolation* VARIAC in order to have the device under test floating. This is a safety feature, so invest a bit. Plenty of choices are available on online auction sites.
Make sure the VARIAC is actually isolating (most pot-shaped VARIACS of US origin are not isolating as far as I can tell).
The VARIAC would also allow you to test any power supply against line voltage changes - many calibration procedures for O'scopes include this test.
Stay safe,
Magnus
|
David Hess
KB6NAX
David C. Partridge