The 'Crest' or 'Peak Hold' function probably works fine if it is selected after the circuit is energized. It should register transient durations of 1mS or longer. Inrush current measurement, with the 'Crest' selection, obviously was not considered by Brymen at meter conception .
The Max/Min feature might detect inrush current, if it lasts for longer than 200mS. Has anyone tried it yet?
The point is if you want to measure a pulse of current that doesn't exceed 0.8 amps, it won't be visible as the Crest function causes the meter to immediately read 0.83 amps. The meter is blind to anything under 0.83 amps when using Crest mode. Yes I have tested this and it gets worse if there is actually a current flowing before you push the Crest button. With one amp, the meter displays 1.79 amps in crest mode at start up.
The 'Crest' or 'Peak Hold' function probably works fine if it is selected after the circuit is energized.
Hm? Not sure what you are referring to here - the meter jumps to 0.85A the moment the CREST button is pressed, regardless of whether the meter is shorted, open-circuit or in-circuit (although I've seen it jump higher if in circuit). The whole point is that it should start at zero before the circuit is energised, but it doesn't. It starts off at 0.85A, rendering any in-rush measurement useless.
It should register transient durations of 1mS or longer. Inrush current measurement, with the 'Crest' selection, obviously was not considered by Brymen at meter conception .
The manual specifically refers to "... or current signal duration as short as 1ms ... available to display 5000uA, 500mA and 10A ranges". I tested my circuit's inrush current using a 0.2 ohm shunt resistor and measured it on the voltage scale. The circuit spikes to 900mA on switch-on. Confirmed this with a DSO over the same shunt resistor and figures are within 5% of each other, which is good enough for me.
So the meter is certainly capable of measuring inrush - the A range must just start at zero.
The Max/Min feature might detect inrush current, if it lasts for longer than 200mS. Has anyone tried it yet?
Yah - tried it a few times, but not really suitable in this scenario - it only detects between 180mA and 250mA (depending on how I catch it) using my test circuit.
--deckert
Anyone have any news from Brymen?
No yet, but I first reported it to Brymen late on Friday, which was already after 5pm in Taiwan, so I'll cut'em some slack till Monday or so
--deckert
Here is the reply from Brymen:
Thanks for feeding back comment to us.
This is not its bug. Actually it is its nature out of its design. BM869 Crest mode design is targeted to capture the peak value short to 0.8ms in duration. To hit this target, very high speed sample rate is in need. In such case, its noise rejection can not be good as normal measurement mode naturally. That is the reason why we add an additional tolerance for CREST mode to manual specs as followings.
Crest mode (Instantaneous Peak Hold)
Resolution: 5000 counts
Accuracy: Specified accuracy +- 100 digits for changes > 0.8ms in duration
Best Regards,
Gary Wang/Brymen
Product Marketing Manager
Yes this specification is in the manual, I somehow missed it. These meters are operating within specification. It is not that useful a function with these very loose specs but it does have its uses.
Yes this specification is in the manual, I somehow missed it. These meters are operating within specification. It is not that useful a function with these very loose specs but it does have its uses.
Agreed, as per spec, the meter is within limits. I will admit that it makes it less useful on the 10A scale, so using a shunt resistor seems to be the only practical way of measuring current in-rush.
I got the same response, but I'm still very pleased with the response from Brymen (from their marketing manager, no less). No real delay in getting an answer from them.
--deckert
Ok, then how can this be explained:
Just did the test on my BM869, didn't get the same result.
- Shorted COM and A
- Turned on to A scale, reading is about 0.0001A DC with 0.000A AC
- Pressed crest: OL (reading is DC only) on both MAX and MIN.
P.S.: Brymen gave me the same answer.
They have a 45 mOhm shunt, the shot noise difference compared to the voltage range measurement with an external 0.2 Ohm shunt is (sqrt(0.045)/0.045)/((sqrt(0.2)/0.2))~= 2 times worse (relatively). It's nice that the meter works in specifications, but their explanation doesn't hold water unless they run the measurement a whole lot slower in the voltage range.
Not a bug, it's a feature...
I see this post is a bit old, but I wanted to tell of my results with my 869. In A mode with Crest engaged I get 37 ma in Min. If I charge a cap an whip the lead from an LED(in series with the meter), I get a max of 507 ma. Yes it would be nice if it read zero for min, but it works great as a max peak, measuring as fast as I can whip a lead past a charged cap.
Seems to be having for the brymen 869s as well. Bummer should have known it earlier before I bought this month ago.
Anybody has any solution to it ?
I have just purchased a new BM867 but I did not know about this thread until just now. I conducted the same tests on my new meter and I confirm the following
Firmware is 869-P9
DCA Max +0.69A
DCA Min -0.64A
ACA Max -0.16A
ACA Min -1.50A
My serial number is 183270401
Perhaps it is answer to the question: "If Brymen BM869s is cheaper ...., why people would still buy Fluke?"
....to paraphrase title of older thread.
Perhaps it is answer to the question: "If Brymen BM869s is cheaper ...., why people would still buy Fluke?"
....to paraphrase title of older thread.
Let's face it most meters are cheaper than Flukes, I have plenty of Flukes but the Brymen is still my go to meter.
Still I can not understand why Voltage measurement Crest is regardless OK..
and can not understand why this "feature" is present for five years..
e.g.
They have a 45 mOhm shunt, the shot noise difference compared to the voltage range measurement with an external 0.2 Ohm shunt is (sqrt(0.045)/0.045)/((sqrt(0.2)/0.2))~= 2 times worse (relatively). It's nice that the meter works in specifications, but their explanation doesn't hold water unless they run the measurement a whole lot slower in the voltage range.
Seems to be having for the brymen 869s as well. Bummer should have known it earlier before I bought this month ago.
Anybody has any solution to it ?
If you read Brymen's nice explanation n this thread, you will see the "bug" is, in fact, a design choice: they chose to make a crest feature available but correctly reported the loss of accuracy in the manual. This is a typical physical limitation of analog to digital converters. Any ADC running at a faster speed will reduce the SNR and consequently its ENOB.
Any ADC running at a faster speed will reduce the SNR and consequently its ENOB.
But other meters don't suffer from it that much... Anyway, if first 100 counts don't make sense, then why showing them at all? This is just misleading.
I tend to think there is a design error somewhere (but it's good it is documented).
I bought a BM869 a couple weeks ago, and I like the meter very much. However, the Crest mode is unreliable. On the V setting, measuring a 4.1 v battery in Crest mode for example, I can get voltage readings as high as 4.7v, or -4.7 when switching the probes around. Usually, the bad reading is the one not currently showing, e.g., MIN when the measurement is made with MAX showing, and vice-versa. Thus, I avoid Crest mode and use REC mode instead. I cannot reproduce these erroneous readings consistently. Repeating the measurement several times in a row usually does the trick though. If you connect the meter to the circuit BEFORE pressing Crest, then it is useable for catching fast transients and less likely to be spooked by noise.
I would much rather have a Fluke-style Hold function that works normally. I don't care about catching fast ms pulses. I use a scope for that sort of thing.
I wonder if the Brymen BM869 also has this issue. I'm thinking about buying one as a second meter.
I suspect that the BM867 and BM869 may use the same main board. I recall seeing a tear down picture of the BM867 and it looked like there was a temperature stencil on the pcb for the mode selector knob where the BM869 has this mode setting. Temperature measurement is not a feature on the BM867 as I understand.
So maybe BM869 owners can also check for this issue.
I just bought a BM869 (5/2019) and yes, the BM869 has the Crest bug. It is otherwise an excellent meter.