Mine exploration, photographs and mining history for mine explorers, industrial archaeologists, researchers and historians Mine explorer and mining history videos on YouTube Connect with other mine explorers on Facebook
Tip: do not include 'mine' or 'quarry', search by name e.g. 'cwmorthin', use 'Sounds like search' if unsure of spelling

Advanced Search
'Sounds like search'
Quick a b c d e f g h i j k l m n o p q r s t u v w x y z
Tip: narrow down your search by typing more than one word and selecting 'Search for all words' or 'Exact search'

Search for any word
Search for all words
Exact search
Tip: narrow down your search by typing more than one word and selecting 'Search for all words' or 'Exact search'

Search for any word
Search for all words
Exact search

Mine Exploration Forum

Jump to page << < 1 2 3 4 5 6 7 8 9 10 11 > >>
Author My Oldham Conversions
Mr Mike

Avatar of Mr Mike

Joined: 09/06/2007
Location: Bury - In The Laboratory

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 14:03:58
Reply |  Quote
When I have got it sorted I will take some snap shots of it in the lab. Problem is I will only be able to drive it at 5A max, as my bench PSU only goes up to that, so just over half power.

However it will be interesting to compare it a 3.2A to the P7, that will be like for like in power input.

--

Mr Mike www.mineexplorer.org.uk
IP: 80.47.43.114
mountainpenguin

Joined: 18/12/2006

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 14:07:16
Reply |  Quote
they need a lot of grunt don't they!
2 18650's aren't enough to power them, they need 4 or 5 before you can safely deliver enough current to max the thing out!
They will also need a lot of cooling!
IP: 194.74.204.101
Mr Mike

Avatar of Mr Mike

Joined: 09/06/2007
Location: Bury - In The Laboratory

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 16:05:30
Reply |  Quote
One thing going for them, is that the forward voltage is much lower than normal LED's, so with a bit of luck you might be able to use a buck topology running from a 3.7V Li-Ion, getting away from having to use 7.2V Li-Ion simplifying the battery pack.



(click image to open full size image in new window)



--

Mr Mike www.mineexplorer.org.uk
IP: 80.47.43.114
mountainpenguin

Joined: 18/12/2006

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 16:14:36
Reply |  Quote
low VF is all well and good but you need more current.
I have managed to make a 4 led system that draws more than 2c so it cant run on protected cells I defiantly wouldn't want to run a caplamp like this.
I think you will still need a driver to regulate the current anyway so I am not sure what benefit you get from running at 3.7V ?
IP: 194.74.204.101
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 16:19:40
Reply |  Quote
The SSC/Cree quads are all low Vf but wont run on single or multi parallel 18650s.

Off the top of my head, hot off charger about 4.16V = 2.2 amps
down to 3.7V = 1.4 amps

Mr Mike has cleverly designed a small compact driver that will drive a quad of single Li Ion.
Have you produced any yet Mike?

--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 92.28.146.142
Mr Mike

Avatar of Mr Mike

Joined: 09/06/2007
Location: Bury - In The Laboratory

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 16:34:14
Reply |  Quote
Just not having to have 7.2V overhead for the driver, so the battery pack is easier to make. If your making 7.2V then you need to balance them with a PCB etc..., just more of a ball ache, whilst if you are paralleling 3.7V cells its easier as you don't have to balance them - but also you can get say a 10Ahr or 20Ahr single cell, making life much simpler.




--

Mr Mike www.mineexplorer.org.uk
IP: 80.47.43.114
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 16:40:54
Reply |  Quote
Right, now I get it.

There lies the advantage of the lower Vf of the LEDs. I didn't understand you at first. This why I have had to have individual drivers to each LED in my maggie, Crees are at 4V or higher.

Looking at it another way, individual circuits does increase the reliability factor, but a right pain to build.

--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 92.28.146.142
Mr Mike

Avatar of Mr Mike

Joined: 09/06/2007
Location: Bury - In The Laboratory

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 17:18:28
Reply |  Quote
royfellows wrote:


Mr Mike has cleverly designed a small compact driver that will drive a quad of single Li Ion.
Have you produced any yet Mike?


No, been snowed under with other projects, I did build the prototype and it worked to a certain extent, but not as good as I hoped (the regulation was bad at about 10% over the input voltage range). A recap to those who may not know what I was doing - my aim was to design a driver for the P7 that would drive it fully at 2.8A from 3.3 to 6V input with a number of power levels, allowing you to use a single 3.7V supply.

--

Mr Mike www.mineexplorer.org.uk
IP: 80.47.43.114
mountainpenguin

Joined: 18/12/2006

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/01/2010 22:40:47
Reply |  Quote
Mr Mike wrote:

Just not having to have 7.2V overhead for the driver, so the battery pack is easier to make. If your making 7.2V then you need to balance them with a PCB etc..., just more of a ball ache, whilst if you are paralleling 3.7V cells its easier as you don't have to balance them - but also you can get say a 10Ahr or 20Ahr single cell, making life much simpler.




ummmm you do have to balance parallel cells ...

IP: 83.67.133.116
Mr Mike

Avatar of Mr Mike

Joined: 09/06/2007
Location: Bury - In The Laboratory

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 16/01/2010 09:21:45
Reply |  Quote
No, with Li-Ion you don't - unless I am really missing something? With NiCD and NiMH you do. I've had a number of 'off the shelf' packs of Li-Ion and they were only protected with an overcharge / under voltage protection PCB, no balancing.

--

Mr Mike www.mineexplorer.org.uk
IP: 80.47.133.79
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 28/01/2010 16:52:40
Reply |  Quote
This is just a repeat of what I have posted to ME, for the benefit of those here who dont use that site.

I have just received my R5 from KD and run some tests

Mr Kai Lam velly reliable on delivery and stock levels, credit where it’s due, but a bit pricey. Still, don’t matter how cheap if non in stock.

Results are as expected.
Picture #1 is a Cree Q5, lamp used for testing was my own UniStar on single lamp full.



(click image to open full size image in new window)

Picture #2 is Cree R5 using exactly the same reflector and AMC 3 chip 1050 mA driver.



(click image to open full size image in new window)

Picture #3 is R5 but using Mr Lams 27.8mm X 14.8mm aluminium reflector S006507.



(click image to open full size image in new window)

Now this is an interesting piece of kit being small enough to fit in an Oldham headset together with a small 18mm reflector from DX.

I have very little doubt that I can get a Q5 to throw as tight a beam as my treasured Fenix TK11 using this reflector. I will experiment with this and also using Cree MCE quad die. The only thing to remember is that it is not everyone that likes tight beams, there has to be a compromise in a caplamp, but I cannot help but think that there may be mileage in using an R5 with this reflector as one of the lamps in a UniStar


--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 78.145.250.19
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 29/01/2010 09:36:01
Reply |  Quote
Last night I did some tests in the dark outside.
The R5 with the 28mm reflector is bloody marvellous, trust me.

With an 18mm reflector its light going everywhere, and worse still used in one of the borings of the 5 bore cast aluminium reflector as used on the MagnumStar.

A UniStar with an R5 would have to have the 28mm reflector and a Q5 with the 18mm as the second emitter and should be an incredible lamp for the money. I shall build one for my own use to evaluate it.

The 28mm lens from KD is very good value for money, and I have ordered 20 of them and 5 R5s.

A Q5 with this lens gives a beam as tight as my Fenix TK11.

I have just run some comparison tests for temperature gain against a Q5

The tests ran for 5 minutes, the start temperature was 10 degrees centigrade; both used the AMC 7135 3 chip driver. The current out to the R5 was 1.12 amps, to the Q5 it was 1.07 amps. I am not comfortable about the Iout to the R5 and will tune it down a bit when I build lamps.

Cree R5 XPG: Temperature gain was slow, about 1 degree every 6 seconds, after 5 minutes it was 31 degrees, I would expect it to max out about 30 to 33.

Cree Q5 XRE: Temperature gain was fast and after 5 minutes it was up to 38 degrees, I would expect it to possibly climb another 5 to 8 degrees before maxing out.

Some thoughts.

It appears that the R5 runs considerably cooler than the Q5 and that one of these emitters could be used in an Oldham headset without external heatsinking. I am aware that people are doing this with Q5s but I am not comfortable with this.
It is my opinion that the R5 is useless however as a caplamp, without the larger reflectors from KaiDomain, these make all the difference and can form the basis of a formidable lamp. These are designed for Q5s however and have a base hole to suit; the R5 has a much smaller footprint so it is important to centre the emitter in the reflector hole.
Obviously, my MagnumStar will remain unchanged as my power beam must be what it says on the tin. The Unistar would definitely benefit from one of these R5s with the larger reflector. By machining the reflector down slightly on the side 2 could be fitted side by side in an Oldham headset, but this would make the screws that attach the LED module to the leg of the thermal conductor inaccessible. Length wise the fit is not a problem, my lamps use replacement 3mm front seals that gives a little extra clearance anyway. The output current can easily be reduced slightly with a series resistor, the value could be very low such as 0.05 ohms so the power wastage is negligible, it merely fools the driver into thinking its outputting more current than it is.

I am producing a revised WorkStar as the 2 bore cast aluminium reflectors are no longer available, so I shall incorporate these reflectors for main beam on my new design that assembles on a copper plate. An advantage with this is that the LED modules can be easily changed.

It is looking highly likely that I will be producing drop in modules for the DIY man, these will use the R5 and 28mm reflector as main emitter, and a low power pilot in a small reflector. If the tests run OK I will post the full details.

Referring back to my postings on M.E. I calculated the light output of the R5 at 1070 mA to be 347 lumens, if I remember correctly.

I do have to say that if the spatial distribution had been similar to the Q5 this would have the same impact on LED caplamps as Windows 95 had on computers way back in the 90s, but its not all over yet by a long way.


--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 78.145.154.126
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 29/01/2010 15:21:05
Reply |  Quote
More

Very roughly.

Vin = 3.5V Iin = 0.9 amps
Vout = 3.2V Iout = 1.02 amps

Watts in = 3.26 watts out = 3.15
Series resistor used with 3 chip AMC 7135 = 0.14 ohms.

Only get from Farnell or similar, 1 watt should be OK.

Surprised about the efficiency, expected good but this is ridiculous.

Obviously longer burn times as well, Q5 on full wack is over 4 watts in.



--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 92.28.171.237 Edited: 29/01/2010 15:22:24 by royfellows
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 03/02/2010 18:43:13
Reply |  Quote
I have done some thorough testing, what I posted before was TOTAL RUBISH and should be ignored.
I don’t know what went wrong, but it was done rather quickly.

The resistors do not make much difference to the efficiency, and the initial Iout drops rapidly to a more acceptable level.

Now where those initial readings came from is a mystery to me I am unable to duplicate anything like them with or without the resistors.

Battery hot off charger gives Iout of 1.11 amps which dropped rapidly, a battery down to 3.68V unloaded gave 1.06 amps out without the resistors, but down to 0.9 amps with them. My opinion now is that the resistors could be dropped.

Efficiency on the new tests is disappointing, being as I correctly stated earlier worse when a battery is hot off charger at 79% getting better up to 86% when the battery falls to below 3.7V unloaded.

These results are disappointing but I would not say unacceptable, some may disagree.

I also tested with a 2 chip AMC driver, this delivered 700 mA as it’s supposed to but at only about 77% efficiency.
A multimode board on the low setting would probably be down to about 60%, we shall have to see.

The reason for the poor efficiency is in my opinion easy to appreciate, its down to the low Vf of the R5 compared with the relatively high Vf of the Q5s. Vout of a Q5 at 1 amp is 4V or more, close to the battery voltage, the R5 is about 3.15V

The thing about this is that lamp builders can use one of these emitters with a 2 chip AMC driver and although only running 77% efficient initially they will only be pulling 2.8 watts from the battery and getting probably more light than a Q5 at 4 watts while generating less heat. Its all roundabouts and swings.

Of course, lurking in the background is the issue of running them of Li Ions in series, I think I can read the tealeaves here and say efficiency would be horrific, but then again pairs would be series wired and this could then change.

More experiments.


--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 78.145.184.120
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 04/02/2010 18:43:21
Reply |  Quote
WorkStar new design

Out of necessity I have had to develop a new version of my WorkStar lamp due to non availability of the cast aluminium reflector unit previously used.
That unit was designed for hybrid LED/Bulb torches so this is not unexpected; I would imagine a demand for these equal to that of Betamax video tapes.

They say that necessity is the mother of invention, and that is what has happened here.
I have taken the simple pre-existing specification of SW1 = Cree Q5 at 700 mA, and SW2 = Pilot LED at 500 mA and mounted it all on a copper plate similar to my UniStar Lamp.

However this goes a bit further.

While the UniStar mounts the copper plate onto the leg of the thermal conductor using 2 X M3 screws, this mounts both top and bottom. The bottom is attached to the positive terminal block using the pre-existing screws, and top into 2 adaptor posts that screw into the holes that took the mounting screws for the pilot bulb housing.

Now the AMC 7135 drivers use a common positive, this is now provided by the mounting plate that acts as positive ground.

So to where is the heat conducted is a very good question.

I have just run one for an hour in my kitchen, the temperature inside the lamp maxed out at 27 degrees, the ambient was 12 degrees.

So where did the heat go.

It as conducted into the mounting bracket, the positive terminal block, and into the copper or brass bus system that is moulded inside the plastic headset. There is no direct metal to metal contact between the bracket and internal parts, and yet it was noticeably warm at the bottom, so it must have took some of the heat.

Of course, here we have a situation whereby the LED module is easily removable and can be supplied as a kit for DIY conversion, the outdoor centre for whom I am converting lamps can keep spares, and in the event of a lamp becoming faulty they can swop the unit and just post the module to me not the whole lamp.

Further implications are a DIY unit that runs the main LED at full power for as long as you like, but I first want to complete my experiments with the r5s. A single R5 with the large reflector could be all the lamp that most would want may well be available as drop in conversion.

Photo below is of lamp with unit as just made up, eventually the innards get painted black.



(click image to open full size image in new window)



--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 92.26.141.143
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/02/2010 12:27:12
Reply |  Quote
CREE R5 Emitters.

Over the weekend I have been testing one of these in a slate mine using an improvised caplamp, and have to give all other lamp builders a warning about these.

In any kind of a reflector the beam cut off is absolute, and is several feet in front of the wearer. As such your feet and immediate foreground are in total darkness, it is absolute.
I can see a situation whereby anyone using a lamp built around one of these emitters without some form of supplementary foreground illumination could get killed or injured by tripping over or walking into something. This is no exaggeration.
I cannot see this situation being rectified by purpose built reflectors when they become available, as these will obviously be designed for torches where foreground illumination is not a consideration.

I have to confess to feeling bitter disappointment.

On impulse I tore the reflector off my improvised lamp and tried it as a bare emitter, it was breathtaking and I can see the cavers going ape over this. I was in a large slate cavern and yet all around me was well illuminated.

Obviously this takes the use of this emitter out of the frame as far as my WorkStar lamps go, any lamp designed around this would have to have a supplementary bare emitter. I can use 3 of these in my MagnumStar and boost the output to over 2000 lumens without a problem, we shall have to see.

I may also try to design a lamp with 3 of these, with one emitter being bare.

Anyway, please take a warning from what I say above.


--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 92.28.161.36
AR

Avatar of AR

Joined: 07/11/2007
Location: Knot far from Knotlow in the middle of the Peak District

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/02/2010 12:47:50
Reply |  Quote
That sounds like a major hitch with the reflectors Roy - I can think of several mines where a lamp that kept your feet in darkness would very likely result in an NRT descent of a winze! Shocked

This is just a spur-of-the-moment idea so feel free to shoot me down in flames on this, but what would happen if you mounted an R5 so that it sat close to the glass of a lamp rather than in the "usual" position? potentially this would allow a polished flat heatsink close to the glass, though whether this would actually do much for lighting patterns and levels I leave to those of you with more experience in these matters.....

--

I sold my soul to Satan, but he brought it back and demanded a refund....
IP: 194.159.145.70
royfellows

Avatar of royfellows

Joined: 13/06/2007
Location: Great Wyrley near Walsall

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/02/2010 14:59:43
Reply |  Quote
YES!
This is exactly what I am thinking of doing. Obviously, the nearer the glass a bare emitter, the more light out at the front. And as I have said, it is a fantastic flood, just like the old carbide light that the Scurion people try to emulate.

But a lamp for all seasons will have to combine a beam with this.

The other consideration is that mounting a bare emitter close to the glass would give room behind it for driver boards etc.

--

''the stopes soared beyond the range of our caplamps' - David Bick...... How times change
IP: 92.28.161.36
AR

Avatar of AR

Joined: 07/11/2007
Location: Knot far from Knotlow in the middle of the Peak District

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/02/2010 15:44:25
Reply |  Quote
So, I take it your next experiment will be one or two LEDs mounted on a flat plate close to the glass to provide spill with another set back in a torch reflector to provide a focused beam. With a large enough headset (say, an old Youle) you might even be able to fit a battery in as well like the KSE headlights.....

--

I sold my soul to Satan, but he brought it back and demanded a refund....
IP: 194.159.145.70
Morlock

Avatar of Morlock

Joined: 31/07/2008

View Profile
View Posts
View Personal Album
View Personal Files
View all Photos
Send Private Message
My Oldham Conversions
Posted: 15/02/2010 15:47:05
Reply |  Quote
I've just tried my R2 conversion combined with helmet and normal lamp bracket angles and can only say the R2 is many times better than the standard Oldham with specular reflector. IP: 82.26.167.18
Jump to page << < 1 2 3 4 5 6 7 8 9 10 11 > >>
Moore Books: Specialist Books I.A. Recordings: Mining and Industrial History DVDs Starless River - Caving Store Explore a Disused Welsh Slate Mine
Disclaimer: Mine exploring can be quite dangerous, but then again it can be alright, it all depends on the weather. Please read the proper disclaimer.
© 2005 to 2015 AditNow.co.uk
Top of Page