V U meter circuit - UK Vintage Radio Repair and Restoration Discussion Forum

Mike Brett · 27th Dec 2012, 7:29 pm

Hi all
Can anybody give me some help with this simple circuit as I have built it and its driving me nuts. I have built it as it is apart from capacitor C1 , I did not have a 3.3 u so I used a 63 v 4.7 u. When I connect the input to an audio signal with VR1 set about half way nothing lights up. If I reduce VR1 to lower the power going to the chip, L E D 1 to 6 light up very dimly and do not alter with the volume. I have tried a radio with this circuit and also a audio signal generator but the results are the same. Looking at the circuit do you think it is possible to work.
Cheers Mike

Mike Brett · 27th Dec 2012, 9:12 pm

Oops:

http://www.eleccircuit.com/wp-conten...-ic-LM3914.jpg

Mike Brett · 27th Dec 2012, 9:42 pm

More observations. The speaker output I am trying to monitor although very loud has almost no dc voltage on it and as I believe the LM3914 needs about 1.2 volts on input pin 5 that could be my problem. Is there anyway of driving this circuit from a small speaker or am I going about this the wrong way.
Mike

Ed_Dinning · 27th Dec 2012, 10:19 pm

Hi Mike, most speakers do noy like DC voltages on them; this may even confuse the VU meter. Try measuring the speaker voltage with you DVM on AC volts, but through a 1uF cap to block any DC. If still low then you may need a small transformer to boost the voltage for your VU meter.

Ed

MrElectronicman · 27th Dec 2012, 11:54 pm

In looking at the application notes on this IC, it looks like it may need at least 3 Volts to register anything.

dave cox · 28th Dec 2012, 1:08 am

I suspect that when you rotate VR1 to minimum that you are just seeing the lowest LEDs oscillating or maybe just the noise in the comparitors. The main problem I see with this arrangement is that each comparitor need a 'minimum' voltage to operate - I don't see this stated in the data sheet but I would guess it's around 120mV as there are 10 steps and the ref voltage is 1.2V (band-gap ref ?). Any position of VR1 that puts less than ~ 1.2V on pin 6 is likely to be problematic. Additionally, as the LEDs get switched there is likely to be coupling back to pin 6 so it's quite likely to oscillate !!

Also, the data sheet doesn't show any series resistors on the LEDs ...

dc

Mike Brett · 28th Dec 2012, 11:08 am

Many thanks for the help,gives me something to work on now.
Cheers Mike

cmjones01 · 28th Dec 2012, 11:13 am

That circuit has one important mistake: pin 8 of the chip needs to be grounded (as well as all the other stuff connected to that pin). In addition, I've found that the LM3914/5/6 benefit from some power supply decoupling (1uF across the supply should do it).

The circuit has a very crude rectifier so won't be very sensitive. An audio meter like this really needs an op-amp based precision rectifier.

From memory, I think the LM3914 has a linear scale, so when used on an audio signal it'll give you pretty flashing lights but will be meaningless for checking levels. I think the 3915 has a log scale, which is useful, and the 3916 has a VU scale - log but with uneven steps so as to have resolution in the right places.

My memory of the part numbers may be wrong, so check the data sheet!

Chris

cmjones01 · 28th Dec 2012, 11:14 am

And by the way, the LM391x series doesn't need LED series resistors. It has constant current drive built in.

MrElectronicman · 1st Jan 2013, 12:52 am

After reading this post I was inspired to dig out a piece of an old cassette player that had a nice digital Vu meter. By today's standards it's a bit homebrew as it uses an A/D converter to drive the LED's and a OP Amp as a compressor to provide the A/D converter drive.

I rigged it up using a 5VDC supply for the LED's and what should have been 21VDC for the OP.Amp. Since I did not have a 21 volt supply handy, I used 14VDC, which worked.

Connected it across the speakers of my stereo amp and got more than enough voltage to drive the whole thing.

It's a bit ungainly as it's on the same PCB as the tape counter and tape control indicators, and would be difficult to rebuild it on another PCB of it's own.

Thanks for the idea, and sorry if I may have hijacked this post

27th Dec 2012, 7:29 pm	#1
Mike Brett Retired Dormant Member Join Date: Dec 2005 Location: Wisbech, Cambridgeshire, UK. Posts: 452	V U meter circuit Hi all Can anybody give me some help with this simple circuit as I have built it and its driving me nuts. I have built it as it is apart from capacitor C1 , I did not have a 3.3 u so I used a 63 v 4.7 u. When I connect the input to an audio signal with VR1 set about half way nothing lights up. If I reduce VR1 to lower the power going to the chip, L E D 1 to 6 light up very dimly and do not alter with the volume. I have tried a radio with this circuit and also a audio signal generator but the results are the same. Looking at the circuit do you think it is possible to work. Cheers Mike

27th Dec 2012, 9:12 pm	#2
Mike Brett Retired Dormant Member Join Date: Dec 2005 Location: Wisbech, Cambridgeshire, UK. Posts: 452	Re: V U meter circuit Oops: http://www.eleccircuit.com/wp-conten...-ic-LM3914.jpg

27th Dec 2012, 9:42 pm	#3
Mike Brett Retired Dormant Member Join Date: Dec 2005 Location: Wisbech, Cambridgeshire, UK. Posts: 452	Re: V U meter circuit More observations. The speaker output I am trying to monitor although very loud has almost no dc voltage on it and as I believe the LM3914 needs about 1.2 volts on input pin 5 that could be my problem. Is there anyway of driving this circuit from a small speaker or am I going about this the wrong way. Mike

27th Dec 2012, 10:19 pm	#4
Ed_Dinning Dekatron Join Date: Jan 2004 Location: Newcastle upon Tyne, Tyne & Wear, UK. Posts: 8,195	Re: V U meter circuit Hi Mike, most speakers do noy like DC voltages on them; this may even confuse the VU meter. Try measuring the speaker voltage with you DVM on AC volts, but through a 1uF cap to block any DC. If still low then you may need a small transformer to boost the voltage for your VU meter. Ed

27th Dec 2012, 11:54 pm	#5
MrElectronicman Heptode Join Date: Feb 2011 Location: Washington DC, USA Posts: 619	Re: V U meter circuit In looking at the application notes on this IC, it looks like it may need at least 3 Volts to register anything. __________________ David

28th Dec 2012, 1:08 am	#6
dave cox Nonode Join Date: Jan 2009 Location: Bristol, UK. Posts: 2,062	Re: V U meter circuit I suspect that when you rotate VR1 to minimum that you are just seeing the lowest LEDs oscillating or maybe just the noise in the comparitors. The main problem I see with this arrangement is that each comparitor need a 'minimum' voltage to operate - I don't see this stated in the data sheet but I would guess it's around 120mV as there are 10 steps and the ref voltage is 1.2V (band-gap ref ?). Any position of VR1 that puts less than ~ 1.2V on pin 6 is likely to be problematic. Additionally, as the LEDs get switched there is likely to be coupling back to pin 6 so it's quite likely to oscillate !! Also, the data sheet doesn't show any series resistors on the LEDs ... dc

28th Dec 2012, 11:08 am	#7
Mike Brett Retired Dormant Member Join Date: Dec 2005 Location: Wisbech, Cambridgeshire, UK. Posts: 452	Re: V U meter circuit Many thanks for the help,gives me something to work on now. Cheers Mike

28th Dec 2012, 11:13 am	#8
cmjones01 Nonode Join Date: Oct 2008 Location: Warsaw, Poland and Cambridge, UK Posts: 2,681	Re: V U meter circuit That circuit has one important mistake: pin 8 of the chip needs to be grounded (as well as all the other stuff connected to that pin). In addition, I've found that the LM3914/5/6 benefit from some power supply decoupling (1uF across the supply should do it). The circuit has a very crude rectifier so won't be very sensitive. An audio meter like this really needs an op-amp based precision rectifier. From memory, I think the LM3914 has a linear scale, so when used on an audio signal it'll give you pretty flashing lights but will be meaningless for checking levels. I think the 3915 has a log scale, which is useful, and the 3916 has a VU scale - log but with uneven steps so as to have resolution in the right places. My memory of the part numbers may be wrong, so check the data sheet! Chris

28th Dec 2012, 11:14 am	#9
cmjones01 Nonode Join Date: Oct 2008 Location: Warsaw, Poland and Cambridge, UK Posts: 2,681	Re: V U meter circuit And by the way, the LM391x series doesn't need LED series resistors. It has constant current drive built in.

1st Jan 2013, 12:52 am	#10
MrElectronicman Heptode Join Date: Feb 2011 Location: Washington DC, USA Posts: 619	Re: V U meter circuit After reading this post I was inspired to dig out a piece of an old cassette player that had a nice digital Vu meter. By today's standards it's a bit homebrew as it uses an A/D converter to drive the LED's and a OP Amp as a compressor to provide the A/D converter drive. I rigged it up using a 5VDC supply for the LED's and what should have been 21VDC for the OP.Amp. Since I did not have a 21 volt supply handy, I used 14VDC, which worked. Connected it across the speakers of my stereo amp and got more than enough voltage to drive the whole thing. It's a bit ungainly as it's on the same PCB as the tape counter and tape control indicators, and would be difficult to rebuild it on another PCB of it's own. Thanks for the idea, and sorry if I may have hijacked this post __________________ David