HP 1740A on the bench, help please?

[tggzzz]

...Seeing as they are only 16 miles from me I offered...

When I'm not sure whether to have something posted, I calculate the travel cost as 20p/mile. That's ~12p/mile for petrol, and 8p/mile for depreciation and insurance. On that basis, 32 miles = £6.40.

[CJay]

...Seeing as they are only 16 miles from me I offered...

When I'm not sure whether to have something posted, I calculate the travel cost as 20p/mile. That's ~12p/mile for petrol, and 8p/mile for depreciation and insurance. On that basis, 32 miles = £6.40.

Yeah, that and the cost of the 'ooh, that's handy' factor which always doubles the spend when you visit somewhere and see their stock.

[Specmaster]

...Seeing as they are only 16 miles from me I offered...

When I'm not sure whether to have something posted, I calculate the travel cost as 20p/mile. That's ~12p/mile for petrol, and 8p/mile for depreciation and insurance. On that basis, 32 miles = £6.40.

True, but economy, that could take 5 working days, I wont to crack on with it, bad enough waiting for the MPSH81's coming from Hong Kong.

I got to thinking that maybe I did this when testing for shorted tants using the Keithley 160 so I did voltage check on it in the resistance ranges and I can say that wasn't the cause of the failing, it only gives out on the lowest range 1.3V and on the 1000M range, a staggering low value of 1.4mV. This then got me thinking out of the box and checked my other bench meters and the results are amazing.

Kikusui 1502              Low range 430mV   High range 215mV

Iso Tech IDM-203       Low range 400mV   High range 200mV

Philips PM2521           Low range  3.8V      Auto  1V

HP 3466A                   Low range 4.6V      Auto 1.1V

So the Iso Tech and the Kikusui seem to be  the best ones in circuit testing.

I wonder how the Fluke 8505A will compare, when it finally arrives, it still sits in Heathrow according to the tracking info.

[Specmaster]

...Seeing as they are only 16 miles from me I offered...

When I'm not sure whether to have something posted, I calculate the travel cost as 20p/mile. That's ~12p/mile for petrol, and 8p/mile for depreciation and insurance. On that basis, 32 miles = £6.40.

Yeah, that and the cost of the 'ooh, that's handy' factor which always doubles the spend when you visit somewhere and see their stock.

To be honest, thats one of the reasons I like to collect if possible, very often these places will have special deals in the shops / stores that they can't advertise because of limited stock etc. Always a chance of picking up a reaL bargain while there. 

[Specmaster]

She lives again, Q1 replaced to remove the incorrect and unidentifiable transistor as shown in q1 picture which had its legs criss crossing in all directions to try and make it fit into the socket and was failing most of the time and also replaced Q17 which reduced the -15V to -12V, or should I say it was supposed to but I got 1.89V on test.

The main power supply rails were rock solid and that was confusing me and after much head scratching and searching about on the various boards I noticed that there was a particular requirement for a + and - 12V rail that was only applicable to vertical section and thats when the lack of the -12V was detected.

The calibration is a mile out however so I'll tackle that another day, but to give you an idea, 1v test square wave will give a 1 Cm division, not on the 1V as expected but on the .1V on both channels so it looks like gain issues there.

I still have the thorny problem however with the undoing of the grub screw on the delay timebase shaft, the actual knob has broken away from the brass boss. This needs to be resolved and also a new knob as well. I suppose in the mean time until I can get it sorted I could use it without, as if it never had a delay feature.

[tautech]

The calibration is a mile out however so I'll tackle that another day, but to give you an idea, 1v test square wave will give a 1 Cm division, not on the 1V as expected but on the .1V on both channels so it looks like gain issues there.

Which is what you'd expect when using 10:1 probes, are you ?

[Specmaster]

Yes I was, I was so engrossed with the repair that I forget scope basics   

[tautech]

Yes I was, I was so engrossed with the repair that I forget scope basics 

Not uncommon user mistake....fingers working faster than brain. 

So really it mightn't be far out of Cal at all, if everything's OK then you only need to verify accuracy not a full Cal that can lead to more gear needed that you probably won't need again.
Traditionally voltage, frequency and mark space references are needed but these can be sourced from a modern AWG albeit not to exacting standards but good enough for the hobbyist.

[Specmaster]

Channel B seems to be OK so far but I'll do more tests tomorrow. Channel A on the other hand displays the 1v test signal as a perfect square wave with input set to .2V and display uses 5 divisions which is spot on but moving onto .5V, 1V etc the trace changes as can be seen in pics. I'm guessing this is a capacitance issue?

Now I know how to take the time base shaft out, I could perhaps remove the shaft and drill the grub screw out if I can then source a replacement knob I could then reinstate the time base knobs (just have to suss out how the knobs react to each with the rachet type of unit that sits between the main and delayed TB knobs. I'll be doing more with this tomorrow.

[tautech]

AC coupling. 

[Specmaster]

AC coupling. 

Yes both channels are set on AC coupling?

[tautech]

AC coupling. 

Yes both channels are set on AC coupling?

This one's not:

[tggzzz]

Traditionally voltage, frequency and mark space references are needed but these can be sourced from a modern AWG albeit not to exacting standards but good enough for the hobbyist.

It is worth comparing the spec of an AWG+DMM with the scope's spec. Which do you think will have wider tolerances?

[Specmaster]

It is, the control is on AC coupling, volts/division is on .2 so the screen shows 5 divisions from the 1v test point, which shows the vertical calibration to ok.

[tautech]

Traditionally voltage, frequency and mark space references are needed but these can be sourced from a modern AWG albeit not to exacting standards but good enough for the hobbyist.

It is worth comparing the spec of an AWG+DMM with the scope's spec. Which do you think will have wider tolerances?

Scope of course. We are talking about CRO's aren't we ?

[Specmaster]

Traditionally voltage, frequency and mark space references are needed but these can be sourced from a modern AWG albeit not to exacting standards but good enough for the hobbyist.

It is worth comparing the spec of an AWG+DMM with the scope's spec. Which do you think will have wider tolerances?

The scope should have the wider tolerance I would have thought?

If you're referring to calibrating it using a AWG & DMM, my AWG is only capable of 25Mhz.  I do however have a Heath kit oscilloscope calibrator.

[tggzzz]

Traditionally voltage, frequency and mark space references are needed but these can be sourced from a modern AWG albeit not to exacting standards but good enough for the hobbyist.

It is worth comparing the spec of an AWG+DMM with the scope's spec. Which do you think will have wider tolerances?

Scope of course. We are talking about CRO's aren't we ?

Precisely - and by a good margin too. Hence I think an AWG would be more than sufficient for any such use, not just hobbyist use.

[tautech]

It is, the control is on AC coupling, volts/division is on .2 so the screen shows 5 divisions from the 1v test point, which shows the vertical calibration to ok.

OK, so the 0V position (no waveform) is where ? In the middle (0V) position on the display ?

If they're all AC coupled then only the first image (0.2V/div) is OK. At .5 and 1V/div there's something wrong.
Houston, we have a problem.

Try the same with DC coupling....which is mostly what one uses anyway.

BTW for amplitude checks I'd just use a known DC source, the SM specifies what's needed.
BTW 2, you shouldn't use the probe cal for amplitude or frequency checks and especially not for internal adjustments, why, well put a counter and another (in cal) scope on it and you'll see why. Probe cal is only for the fast square edge to adjust probes with, the frequency and amplitude are only set to nominal values, albeit that in some scopes it can be adjusted in both amplitude and frequency close to intended spec.

[tautech]

If you're referring to calibrating it using a AWG & DMM, my AWG is only capable of 25Mhz.  I do however have a Heath kit oscilloscope calibrator.

Of greatest value is mark space checks, these are for the timebase accuracy...the part where a scope has its most precision. Ideally you want one that can go close to the BW of your scope in Pulse mode were very short markers at defined periods allow checks of timebase accuracy and subsequent adjustments ...or not.

I used a term a few posts back....verify. That's all I'd do for now as it's probably bugger all out.

Watch this to see Dave do a few minor tweaks on this Hameg:
https://www.eevblog.com/2013/08/03/eevblog-502-19-hameg-analog-oscilloscope/

[Specmaster]

With it set on DC coupling there is a significant offset of the display on both channels as you would expect.

With coupling set to ground then the trace position will vary according the type of coupling, AC with square wave centralized vertically then the trace is on the centre 0 line, on DC the trace when set to ground is on the lowest part of square wave, again as I would expect.

All of the photos are of channel A, the 1st is with volt/div on 0.2V, 2nd photo is with v/d on 0.5V and 3rd is v/d on 1V. The voltage at the test point has been verified with a couple of DMMs to 1V so the screen display is about correct in that case as far amplitude goes.

[tautech]

With it set on DC coupling there is a significant offset of the display on both channels as you would expect.

With coupling set to ground then the trace position will vary according the type of coupling, AC with square wave centralized vertically then the trace is on the centre 0 line, on DC the trace when set to ground is on the lowest part of square wave, again as I would expect.

All of the photos are of channel A, the 1st is with volt/div on 0.2V, 2nd photo is with v/d on 0.5V and 3rd is v/d on 1V. The voltage at the test point has been verified with a couple of DMMs to 1V so the screen display is about correct in that case as far amplitude goes.

That all sounds good but the 0.5V and 1V/div waveforms seem wrong even for AC coupling but it's a while since I used a 1740 or any CRO for that matter. Am I that out of practice ? 

Have you found timebase knobs ?
Sphere wreck these, they must have some tucked away for sure.

[Specmaster]

With it set on DC coupling there is a significant offset of the display on both channels as you would expect.

With coupling set to ground then the trace position will vary according the type of coupling, AC with square wave centralized vertically then the trace is on the centre 0 line, on DC the trace when set to ground is on the lowest part of square wave, again as I would expect.

All of the photos are of channel A, the 1st is with volt/div on 0.2V, 2nd photo is with v/d on 0.5V and 3rd is v/d on 1V. The voltage at the test point has been verified with a couple of DMMs to 1V so the screen display is about correct in that case as far amplitude goes.

That all sounds good but the 0.5V and 1V/div waveforms seem wrong even for AC coupling but it's a while since I used a 1740 or any CRO for that matter. Am I that out of practice ? 

Have you found timebase knobs ?
Sphere wreck these, they must have some tucked away for sure.

No you're not out of practise as these 4 shots will show you, the 1st is B channel DC @ 0.5V, 2nd is B channel DC @ 1V, 3rd is A channel DC @ 0.5V and the 4th is A channel DC @ 1V. This time I have taken the whole fascia so you can see which channel is in use and the settings of the Volt/Div and the coupling mode. Like I said, B channel does seem to be to working almost spot on as far as this test goes, there is some movement on the screen between settings which I can eradicate later by doing the calibration procedure as there are different gain settings etc for different ranges internally that I can tweak.

Whats puzzling me is the odd waveforms on channel A, regardless of the coupling mode, which appear on the higher voltage inputs.

I have seen that video by Dave before but the only adjustment he made was to the trace rotation, the rest was cleaning of contacts and repairing the Auto/Normal trigger button and the tightening, although he did not show it on screen of the vertical verniers as both had become inoperable and he pulled out the B channel knob and shaft. Thats something I can relate to with Hameg scopes as I have a HM408-1A that I'm attempting to repair (both of those had come undone due to broken flexible couplings breaking) but that is a little bit more of a sick puppy compared to the 1740A.

[Specmaster]

No I have yet found a source of the larger of the 2 time base knobs yet, that is the delayed one that the shown brass boss has become detached from, infact the actual knob split into two while trying to undo the bloody grub screw and I just cannot find a screw extractor that goes down small enough to get a bite on the screw head.

I have heard people mention Sphere before but I have no idea where they are located or contact details for them, do have any information on them please?

I really do appreciate your help with this as well by the way, I'd be much further away from completion without your assistance thats for sure. 

[bd139]

Sphere is a bit of a buggering. They’re in Canada. The last thing I got quoted by them which was an HV multiplier for a Tek 465b it was cheaper to buy another unit off eBay for £70 by the time you add shipping, import and Royal Mail to it. The entire scope worked better than the one I was repairing so parts mule duty got swapped over.

This is why when I buy a scope I buy two of them

[Specmaster]

Sphere is a bit of a buggering. They’re in Canada. The last thing I got quoted by them which was an HV multiplier for a Tek 465b it was cheaper to buy another unit off eBay for £70 by the time you add shipping, import and Royal Mail to it. The entire scope worked better than the one I was repairing so parts mule duty got swapped over.

This is why when I buy a scope I buy two of them

I can't buy 2 of them, they are lining up on my workshop (boxroom) floor already and the Hameg (also sick) is in my sons bedroom, SWMBO would be showing me the door if was to a buy another with first getting rid of at least one of the others