Help with analog scope calibration

[o-scope]

I'm new to this forum so hi all! Someone recommended me this place to ask about oscilloscopes and electronic stuff.

I got this old Hitachi V-151B 15Mhz, 1 channel analog oscilloscope for learning purposes and to use it on my old 8-bit computer collection.
I know very basic electronics, I've made some circuits before, I've repaired some stuff, etc.

After playing for a couple of days with the scope I was able to learn how to use it, how to calibrate and what most of the knobs do.
It came with a user manual and a probe.
After reading it, I was able to make a couple of adjustments internally, for example, it wasn't displaying the voltage correctly, on the .5 calibration point it was displaying about 4.6v. After following the adjustment procedure in the manual it nows displays the voltage right. I've tested some batteries and it's working fine.

Now the thing is that I have 2 more adjustments that I can't make, the manual doesn't say anything about these adjustments.
1-Probe compensation.
2- Frequency/period not displaying right.

For the probe compensation I've tried the little trimmer in the probe, but I'm not able to make the flat part of the square totally flat, there's always over or under compensation, I've made a video showing this.

For the frequency/period problem, for example, when I try the 0.5v 1khz calibration signal, if I made the calculation it gives me 1111 Hertz.
I've also tried a phone app with a signal generator set to 1khz but the reading/calculation gives me the same thing.

Now, I'm not sure how accurate this oscilloscope is, but I think that's too much inacuracy right?
Also I'm not sure how reliable these signal generator app really is, but the calibration point on the scope isn't accurate then?

I'm sure that adjusting some of the internal pots I'll be able to calibrate it better, but I'm not sure what to touch. (I'm aware of the high voltages inside)
Any guidance will be very appreciated!!

[james_s]

It's only a 15MHz scope so you will probably not be able to get the square wave to look perfectly flat and square, the rise time is just not fast enough to avoid some rounding.

Hard to say about the frequency accuracy, the calibrator output is only approximate and the phone is probably even less accurate. I would suggest finding one of those little quartz oscillator cans rated for something in the range of 1-10 MHz and see what the scope says about that. They are plenty accurate for calibrating a basic analog oscilloscope.

[w2aew]

As stated in the previous reply, the probe cal signal is there mainly to aid in compensating the probes.  It is likely not going to be a precise calibrated amplitude or frequency - it is not intended for that.

It is possible that the probe can't be fully compensated because the probe doesn't include enough range to compensate for the input capacitance for your particular scope.  Also, a very long ground lead can contribute to overshoot.

As you suspected, very basic entry level scopes like this are likely not going to be terribly accurate, and may not hold their calibration across settings, especially considering it is probably 25-30 years old or more.  Also, make sure that the vertical and horizontal vernier controls are clicked into their Cal positions.

Scope calibration, even for a simple scope like this, can get involved. I have attached two videos that might be helpful.  One talks about 10x probes and the compensation process.  The other is a video showing the calibration process for a basic 25MHz 2 channel analog scope, so that you can see what is involved.

[rstofer]

Are you using a 10x probe?  Usually, the probe compensation only works on the 10x setting.
Given a 10x probe, a 5V source would only measure 0.5V unless the scope has a way to deal with the probe setting.  My DS1054Z does, my Tek 485 doesn't unless I use Tek probes.

[macboy]

Probe compensation trimmer is used for low frequency compensation only. You need to adjust so that the last 2/3 of the square is flat (flat trailing edge). You can't compensate the front, leading edge with this adjustment. Higher end probes will have mid and high frequency compensation to adjust that leading edge too. In your case, the probe/scope combination is a little under compensated at high frequencies, causing a little overshoot. 

[o-scope]

Are you using a 10x probe?  Usually, the probe compensation only works on the 10x setting.
Given a 10x probe, a 5V source would only measure 0.5V unless the scope has a way to deal with the probe setting.  My DS1054Z does, my Tek 485 doesn't unless I use Tek probes.

Yes, I'm using the original hitachi probe X10. Calibration point is 0.5v.

Scope calibration, even for a simple scope like this, can get involved. I have attached two videos that might be helpful.  One talks about 10x probes and the compensation process.  The other is a video showing the calibration process for a basic 25MHz 2 channel analog scope, so that you can see what is involved.

That 2nd video is really useful, as the guy opens the scope to tune it, I'm checking it, thanks!

It's only a 15MHz scope so you will probably not be able to get the square wave to look perfectly flat and square, the rise time is just not fast enough to avoid some rounding.

Hard to say about the frequency accuracy, the calibrator output is only approximate and the phone is probably even less accurate. I would suggest finding one of those little quartz oscillator cans rated for something in the range of 1-10 MHz and see what the scope says about that. They are plenty accurate for calibrating a basic analog oscilloscope.

In the manual it shows a diagram (see pic attached) in which it's suppossed to be able to get the square wave pretty flat, maybe it's the probe? I'm gonna get another probe to test it. Anyway, as I adjusted the vertical gain I was thinking that must be another pot for adjusting the horizontal "gain"?
Also, can you give me a link for those oscillators cans you talk about?
Thank you.

[james_s]

That drawing is just an illustration, you should be able to get it close, but the leading edge is probably never going to be perfectly flat. Be careful making adjustments, many times adjustments interact and must be set up in a specific order, and if you are not positive the instrument is in fully working condition you can make it worse by trying to adjust out a fault.

Proper probe grounding is important, you might want to have a look at AN47 or AN72 from Linear Technology, these focus on higher frequency stuff but even with a few MHz some of the issues can start to show up. One of your screenshots shows a classic case of ringing.

The oscillators I refer to are similar to this:
http://www.ebay.com/itm/2pc-8-MHz-Full-Size-5V-TTL-Crystal-Oscillator-VF150-8-000000MHz-Valpey-Fisher-/121916330914?

They are available in a wide range of frequencies from less than 1MHz to over 100MHz and cost only a few dollars each. Simply hook up 5V to the power pins and a signal appears on the output pin.

[o-scope]

Thank you James, I really appreciate your help