Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
Ripple on Tek 2235 and 2215A
|
Didier Juges
I have a couple of questions regarding Tek 2215A and 2235. I have one each of these and they both have a similar problem: ripple on the trace that gets more visible at high sensitivity (2mV/div). It is barely visible on the 2215A (0.2 division maybe) but it is about 1 division (at 2mV/div sensitivity) on the 2235. It looks like power supply ripple (forgot the exact frequency, but it's in the range of 10-50 kHz AFAICR), but I checked the supply voltages on the 2235 and they all look excellent. I concluded the ripple must be induced some other way but did not manage to find the sneak path.
It does not depend on the input selection (ground, AC or DC) or the position on the screen, only the vertical gain seems to affect the amplitude. Any suggestion welcome. Other than that, these instruments look great and the tubes have great brightness, so I would like to get them working as well as they look. Thanks in advance, Didier KO4BB |
|
Check the spring clip that grounds the crt shield at the top rear of the shield is making good contact. It depends on the tension of the spring clip only to complete the ground path and the contact area is quite small.
toggle quoted message
Show quoted text
Ben At 12:18 24/06/2007, Didier Juges wrote:
I have a couple of questions regarding Tek 2215A and 2235. I have one --
No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.472 / Virus Database: 269.9.6/863 - Release Date: 23/06/2007 11:08 |
|
tom jobe
This question about the small ripple on the 22XX models such as the 2213A,
toggle quoted message
Show quoted text
2215A, 2235, and 2236 is one that I have been wondering about for some time. For my simple purposes this problem is of no real concern, it is more of a curiosity Recently Chuck Harris gave me a good explanation of this problem, and hopefully Chuck will respond to this thread with his experiences. This problem caused him a lot of grief years ago, so he has not forgotten what happened! Hopefully others will offer their experiences and opinions on this issue too. For anyone who wants to check their oscilloscope for this problem, remove the probes, set the inputs to 2 mV/div (at 1X), set the time to 50 uS/div, select either AC or DC for the input (the problem wouldn't happen with the input set to GND), then adjust the "Level" to get the best trace you can get. I just tried this on a very nice 2235, and the square wave ripple was 0.2 division P to P, with a frequency of about 47uS / 21 kHz on both channels. tom jobe... ----- Original Message -----
From: "Didier Juges" <didier@...> To: "!TekScopes" <tekscopes@...> Sent: Sunday, June 24, 2007 4:18 AM Subject: [TekScopes] Ripple on Tek 2235 and 2215A I have a couple of questions regarding Tek 2215A and 2235. I have one11:08 AM |
|
Didier Juges
In my case, the ripple is not a square wave, and it's there regardless of the input setting (AC, DC or Ground).
toggle quoted message
Show quoted text
On the 2235, the ripple amplitude on both channels is different, CH 1 is the worst, even though the channels appear to be about right with respect to calibration. I need to try Ben's recommendation about the CRT shield. Didier tom jobe wrote: This question about the small ripple on the 22XX models such as the 2213A, |
|
tom jobe
Hello Didier,
toggle quoted message
Show quoted text
I wonder if there are multiple problems here? The 2 mV/div setting is not something I use often, and it seems to me that sometimes you can get a "beaded" trace that you cannot stop at the more common volts/div settings. Turning the level knob will sometimes change the direction the "beads" are going along the trace, but you can't seem to trigger on them and freeze the display. Also, these trace "beads" will come and go randomly, and are never very tall P to P. They may be a little wider than the trace, but certainly not a division tall like you have seen. Can you get a small square wave at 2mV/div and 50 uS/div? Hmmmm... tom jobe.... ----- Original Message -----
From: "Didier Juges" <didier@...> To: "!TekScopes" <tekscopes@...> Sent: Sunday, June 24, 2007 6:23 AM Subject: Re: [TekScopes] Ripple on Tek 2235 and 2215A In my case, the ripple is not a square wave, and it's there regardless2213A, time.2215A, 2235, and 2236 is one that I have been wondering about for some aFor my simple purposes this problem is of no real concern, it is more of whatcuriosity removehappened! uS/div,the probes, set the inputs to 2 mV/div (at 1X), set the time to 50 theselect either AC or DC for the input (the problem wouldn't happen with 0.2input set to GND), then adjust the "Level" to get the best trace you can channels.division P to P, with a frequency of about 47uS / 21 kHz on both 11:08 AMtom jobe... |
|
That's interesting, on both previous occasions when I have fixed a ripple problem by cleaning that contact it was Channel 1 that was affected much more than Channel 2. In fact in one case it was only Channel 1, Channel 2 appeared to be perfect. I am at a loss to explain why this should be so ?
toggle quoted message
Show quoted text
Ben At 14:23 24/06/2007, Didier Juges wrote:
On the 2235, the ripple amplitude on both channels is different, CH 1 is --
No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.5.472 / Virus Database: 269.9.6/863 - Release Date: 23/06/2007 11:08 |
|
Didier Juges
The contact was not making very strong pressure, so I cleaned both the
toggle quoted message
Show quoted text
contact and the shield. While the instrument was on, I even put my finger on it and pushed with no difference. Interestingly, the CH 2 ripple is less than the CH 1 but otherwise looks the same shape, but the CH 2 gets a little smaller when the CH 2 switch is put to Ground, but that has no effect on CH 1. Here is a picture of the ripple (20uS/div): Didier KO4BB Ben Clarke wrote: That's interesting, on both previous occasions when I have fixed a ripple |
|
Chuck Harris
A long time ago, in a place far away, and with a brand spanking new 2236
toggle quoted message
Show quoted text
I ran into a problem similar to this. Here is what happened: I had recommended a customer buy a new 2236 over a used 465 or 475 scope. The Tektronix name was compelling, and the price appeared to be an out-and -out steal, $2650 all tricked out. My customer had been using the scope for about a week when he called me up with a problem that he just couldn't figure out. There was a weird train of glitches, steps and spikes riding on a logic signal for a stepper motor controller. He found the strange behavior when he was in the mode of: I have a new scope, lets find something to measure.... So, I came out and looked at the pattern and tried shifting probe grounds and other things, but couldn't make the "ripple" go away. Out to my car I went, and brought in my fancy shmancy 2465 to take a look... The ripple was still there, but now it was about 1/10th as big, and a bit sharper because of the increased bandwidth.... Hmmm? Got smaller with a higher bandwidth scope? That sure looks like an instrumentation problem. We fiddled around with the problem for the day, and I was pretty well convinced that the problem was related to noise going up the power cord and creaming the preamplifier section of the scope. So, I brought in a dusty old 585A scope (also a 100MHz scope) and amazingly the "ripple" signal was totally gone! How could that be??? A 20 year old tube beast scope could do a better job of looking at a signal than a brand new just out of the box Tektronix 2236? To make matters worse, the 585A was able to show a fast, low duty cycle waveform bright as day, when the 2236 needed the room lights turned off with the intensity cranked up to full. My customer was rather disappointed about this to say the least! Up to this point, he really liked his new scope. [My memory is a little fuzzy here, but I have a distinct recollection of finding a note somewhere in the 2236 manual that talked to power line noise rejection. I remember it being something small like 20db. I am pretty sure there was a similar spec in my 2465 manual, but I cannot find it right now. It is possible I got this info from talking to a Tek FAE.] What was happening is the power supply of the stepper motor being tested was dumping noise (that happened to be semi-synchronous to the stepper motor) onto the power line. The noise was going up the cord of the 2236 (and 2465) and through its power supply, and getting into the vertical pre- amplifier section. There it was showing up as a glitchy ripple on the trace. The cure was to either power the scope from another circuit in the lab, or to use a good isolation transformer. In the rush to miniaturize highly capable scopes, we gained something, and we lost something. Something that most people would never notice until long after the sale was completed. -Chuck Harris Didier Juges wrote: I have a couple of questions regarding Tek 2215A and 2235. I have one each of these and they both have a similar problem: ripple on the trace that gets more visible at high sensitivity (2mV/div). It is barely visible on the 2215A (0.2 division maybe) but it is about 1 division (at 2mV/div sensitivity) on the 2235. It looks like power supply ripple (forgot the exact frequency, but it's in the range of 10-50 kHz AFAICR), but I checked the supply voltages on the 2235 and they all look excellent. I concluded the ripple must be induced some other way but did not manage to find the sneak path. |
|
Didier Juges
I can clearly see the ripple going up and down as the sensitivity is changed. I believe the attenuator only provides attenuation in 10x steps, and the other steps are performed by changing the gain of the amplifiers. The ripple is minimum for all scales that have 5 (5V/div, 0.5V/div and so on) and goes up when sensitivity is increased from there down to the 2x and 1x scales, then goes down for the next 5x scale (if that makes sense).
toggle quoted message
Show quoted text
The 2mV/ scale is worse because there is no 5mV/ attenuator, I think the 5mV/ and 2mV/ scales are made with the attenuator in the same 50mV/ setting as the 20mV/ and 10mV/ settingsand just increasing the gain of the amplifier more, which affects the bandwidth and noise. So it looks like this noise must be fed to the front end of the vertical amplifiers. I need to check that the front end has all the shields and grounds in place. Maybe a ground return has been busted, which would not be unusual (the scope was being scrapped by my employer, and we make all sorts of power supplies here, with the usual accidents and smoke, sometimes spectacular) Didier KO4BB tom jobe wrote: Hello Didier, |
|
Didier Juges
I will answer my own question... AA4DF has the 2235 service manual for $12 via FTP and his manuals are excellent. So I am done with that.
toggle quoted message
Show quoted text
Didier Didier Juges wrote: That brings another question: |
|
Greetings!
toggle quoted message
Show quoted text
Sorry to dig up such a historic post, but I have exactly the same problem. The waveform even looks just like Didier's photo. Interestingly, it has a cycle of about 47 uS as mentioned by someone else (although in that case the waveform was a square wave). Has anyone sorted this out? It was mentioned that this scope has poor power supply noise rejection. I did try adding both channels 1 and 2, and inverting channel 2, and this did seem to help quite a bit. It seems like such a hack, however. On an unrelated note, I am feeding the scope with a low-noise preamplifer, and it appears that the preamp is a noise magnet: Just plugging it in, unpowered, causes all sorts of high frequency garbage to appear. Turning on bandwidth limiting helps, but it does not go away. I fear that my home is awash in RF. I'm only interested in signals up to a couple hundred kHz. Is there a way to further decrease the scope bandwidth? Maybe I need a slower scope :) Thank you for any assistance! -J. P. --- In TekScopes@..., Didier Juges <didier@...> wrote:
|
|
For noise issues w scopes take a look at Jim Williams apnotes on the Linear Technologies website www.linear.com
toggle quoted message
Show quoted text
I use a pomona banana plug to connect the low noise amplifier to scope ground (2465). One of the pleasures of the old 5xxx series is their low noise, 'specially w a 5a22n PI. Jack --- In TekScopes@..., "grenert" <grenert@...> wrote:
|
|
It turns out that I am using a preamp very similar to the one in that appnote (AN-83). For those interested, it is here:
toggle quoted message
Show quoted text
I don't quite understand what you mean by connecting the amplifier to scope ground. Could you please elaborate? The preamp assembly directions call for isolated BNC jacks. I wondered if that was the best option, so I "un-isolated" them by soldering a wire from the signal ground to the case. It got rid of _a lot_ of the noise, though not all of it. I wish I had room for something like a 5000 series! Could a passive RC filter (or maybe an LC T-filter) on the output of the preamp serve as a surrogate for BW limiting? I am not interested in anything above about 1-200 kHz. -J. P. --- In TekScopes@..., "caveteursus" <jdwalton@...> wrote:
|
to navigate to use esc to dismiss