Multimeter HP 3457A - problem with fuse blowing

[Nick Gammon]

Hi Martin,

Thanks for your very interesting video about the HP 3457A.  I ordered and received one from eBay recently. It performs as you describe, and I am happy with it ... except for one thing.

After about two days, the (mains) fuse blew when I was turning it on. I hoped I hadn't got a dud, but after I searched for the meter model I found a number of "HP Bench Briefs" describing a similar problem. In particular, the meter may blow a fuse if operated under 220V (as I am, and as I believe you are).

Unfortunately, I can find the summary document (which gives a Bench Brief number) but not the actual document which describes the fix.

There is: MR 11759C-01  Modification prevents fuse blowing in 22OVAC operaton (sic), Document ID: 5846
And: MR 11759D-01  Modification prevents fuse blowing in 22OVAC operation, Document ID: 5847
Also: 3457A-9. Serials 2703A05068 and below. Modification to prevent turn-on failure.

• Have you had this problem?
• Have you solved it?
• Do you (or anyone) know how to obtain these Bench Brief documents which describe the fix?

I installed a new fuse and that hasn't blown yet. Here's hoping. Also, do you think it wise (or unwise, perhaps) to install a slow-blow fuse to try to mitigate the problem? Or is that just asking to have the house catch fire?

[SeanB]

Most likely the fix is the use of a slow blow fuse of the same current rating, along with application of a small label to the rear panel detailing the new fuse value, along with an addendum in the manual detailing this. Probably this was implemented in later models.

[MJLorton]

Hello Nick,

Congratulations on the purchase, I'm sure once the fuse problems are resolved you will really enjoy this "old beast".

Fortunately I had no issues with the fuse blowing.

I've attached the manual and in section 1.5 they talk about using a 80 mA T fuse for 220 / 240v operation. I did use the 200 mA T fuse (meant for 110v) for few days when I first received the unit but then changed it out for a 80mAT fuse. This is a slow blow fuse (as per Sean's comments) as I recall and will hopefully solve your problems.

Note: This would not post with the attachment as it's quite large...I can always email it to you if you don't have it.

Cheers,
Martin.

[Nick Gammon]

Hi Martin,

Thanks for the response. I was uncertain what the "T" meant in 80 mAT, and if you assure me that it means "slow blow" then I will get one, although I could only find 100 mA slow blow on the Element14 web site.

BTW, I don't suppose you have worked out how to use the "memory recording" part of it? I found that part of the manual confusing, to say the least. I sort-of gather that the NRDGS "trigger" setting can be set to take X readings, I'm not sure if that is relevant or not.

I've ordered a GPIB to USB adapter board, so we'll see, when that arrives, if it can usefully be used to take sequential readings.

It would be nice to have it take (say) a reading per hour from a battery as it slowly discharges, and log them to a file.

I already have the manual, and the service manual, thanks for the offer.

I have been working on measuring low-power consumption on the Atmega328P chip in sleep mode, and measuring down to the 100 nA level is useful in those situations.

- Nick

[Monkeh]

T means time-delayed, more commonly known as slow-blow. A 100mA fuse should be close enough, I wouldn't be surprised if the 'modification' is the replacement of an 80mA fuse with a 100mA or 120mA fuse. Increasing the voltage may reduce the operating current, but it increases the inrush current, hence the fuse blowing.

[Nick Gammon]

Thanks for the clarification. As you might imagine, Googling for "maT" returns lots of other responses that are not particularly relevant. Even "fuse mat"

[SeanB]

A slow blow fuse is a compromise for transformers. They will blow almost instantly (  within a single cycle) if subjected to a massive overcurrent ( more than 20x the rating) but will take a second or ten to fail at 5x the current, the timer it takes the element to heat up to melt the fusible link in it. Thus for a transformer where the inrush current is typically 3-5 times the steady state current for two to three complete cycles it will not blow easily unless there is a fault that draws a high current for a while. A standard blow fuse will typically blow at 5x the rating in under 5 cycles roughly, so there you see the reason for the random failure.

Fuses also age, and will often fail open circuit after many years even if not subject to overcurrent spikes. I have 2 400A fuses at work that did this after 40 years of use, though they probably failed from running at 400A for days on end in winter with people cooking in the evenings.

[Monkeh]

Not just transformers, switch-mode supplies exhibit severe inrush, too.

[MJLorton]

Hi Martin,


BTW, I don't suppose you have worked out how to use the "memory recording" part of it? I found that part of the manual confusing, to say the least. I sort-of gather that the NRDGS "trigger" setting can be set to take X readings, I'm not sure if that is relevant or not.

I've ordered a GPIB to USB adapter board, so we'll see, when that arrives, if it can usefully be used to take sequential readings.

- Nick

Hi Nick,

Yes, I did get the memory recording sussed....I can't remember if I showed that in the video...I thought I did. It's not very intuitive so I would have to go back and do some head scratching as I've not done that for a while.

Let me know if you are stuck and I will post the procedure for doing it.

Hats off to you for ordering the adapter board...that will make a great addition.

Cheers,
Martin.

[Nick Gammon]

Just a heads-up. I got a GPIB interface from eBay. It sort-of worked but had a very difficult (impossible?) to use interface. Putting that to one side I then ordered a GPIB-USB Controller. They are available from Sparkfun, amongst other places. Maker: http://prologix.biz/

This was much better. I just plugged it in, opened a terminal program, and typed in (after a bit of head-scratching):

++addr 22

That set the GPIB address to 22 from the default of 5.

Then I could just type in commands from the user manual, and had the satisfaction of hearing relays click as it switched from one mode to another.

Plus, by default it seems to echo each reading as it is taken, so for example, after hooking up a 5K thermistor, and doing this:

Code: [Select]

OHM
RANGE 4E3
MATH CTHRM
I see a nice list of temperatures scroll by:

Code: [Select]

23.124054E+00
23.104186E+00
23.083528E+00
23.052979E+00
23.036073E+00

[MJLorton]

Aah...thanks, very good to know. I will keep that in mind if I venture down that route.