PMEL Forum

Okay, I have a Datron (Wavetek) 4910, 4 channel DC reference standard in my oversized "zener" collection.  As I'm getting ready to start taking stability data on 4 732A's, 1 732B, and 1 Datron 4910, as they are all coming out of mothballs, I'm putting new batteries in all of them.  I've got batteries spec'd out for the 732A's and 732B (pretty low priced).

The real pig is the Datron 4910.  It uses 15 each Powersonic PS-618's (which are obsolete).  This normally shouldn't be a problem, but to fit the foot print, and get the 6V and 2AH rating, I need either a PS-618 or its replacement.

Problem is that all the replacements for the PS-618 are rated at 6V/1AH. 

I'm still looking.  But does anyone know if I can either get by with 1AH batteries, or is there a more creative way to replace them (i.e.: doubling up, etc.).

I even sent an email to Powersonic to request their ideas.

Those are the same dimensions (2 x 2-1/8 x 1-5/8) and power output (1 AH) as Power Sonic PS-610s... I put Powersonic PS-610s in my Datron 4910, and they work fine (just not as long, but we plugged it into a UPS box in case the electricity goes out). But since they're smaller, I had to wedge foam in between the batteries to keep them from rattling around inside. If you do find a battery with the same dimensions as the 618s, let me know.

The one I found that looks like MAYBE it will work is the Peak Energy Products PK6V2, which is a 6V/2AH.  The dimensions are 2.95" (L) x 2.01" (W) x 2.09" (H); and height with terminals is 2.28".  There is one at upsforless[dot]com for $38.50.  I may continue to shop around some more to see if anyone else carries those for a better price.

The other act of desperation was to search at alibaba[dot]com.  If you aren't familiar, it is a website showcasing the actual OEM's products from many Asian countries.  A lot of stuff that is private labelled in the U.S. shows the actual OEM (fascinating site if you have time for the searches - which I don't).

DISCLAIMER:  This is not an advertisement for the above.

I'd be curious if you want to take a look at it to see if it also looks like a go.

My biggest problem is that the batteries for my 732A and 732B are both way cheaper.  If this is what is needed, I can get it.

But it does sway my thinking some that it looks like the 1AH replacements for the PS-618 will at least work.  As I have 4 732A's, 1 732B and 1 4910, I have the advantage that likely, I won't ever have to ship out the 4910.  So maybe the 1AH will be fine, and their only use will be when there are power bumps.

For the 732B try Power Sonic part number PS-1270F1 (12V 7A). 

For the 732A try Power Sonic Part number PS-640F1 (6V 4A).

The battery for the 732A should work in the Datron as well.

In my Datron 4910, there are two layers of interconnected batteries (15 total).  Unless the 4910 was changed a little to allow the taller batteries, it at least doesn't look like I could get away with substituting the NP4-6 (about twice as tall as the PS-618).  They have a strange configuration I may end up schematicizing to put in a substitute (unless I decide to spring for the PK6V2's).  According to the manual, it is a configuration of 3 banks of 5 (with 5 in series = 30V), and paralleled to triple the Ampere Hours (so I guess 30V at a total of 10 AH.  So I may see if I can fit a set of appropriately sized batteries... or maybe I'll just order something.

But I'm wondering if you have some insight to making an NP4-6 (or equivalent) fit in the footprint.  The L and W are similar, but the H is about twice as tall.  Could be I could find a way to remove the separator between top and bottom layers (of existing PS-618's) and put in an array of NP4-6's (4 AH each).

I've learned how this battery system works.  As we all know, they are virtually ALL made in Asia (a LOT in China).  The mark up is about 10X (~1000%).  One battery that after extensive shopping, I found for $38.50, goes for about $3.50 each plus shipping when bought from the factory in Hong Kong.  I am in process of getting a full quote.  But what was going to cost me a total of approx. $1000 for a bunch of lead acid batteries for my array of DC references, is going to cost about $100 including shipping (maybe just slightly more), when I buy from the factory in Asia - which I can do.

I'd be curious to hear info from any of you who have bought parts directly from Asia.

I finally got my battery shipment in from China.  The product is fine and low cost.  As I learned, most all SLA (Sealed Lead Acid) batteries are made offshore - best I can determine, there are no OEM's for most SLA's in the US any more.  What they do is private label, etc.

If I had it to do over again, I think I'd prefer not to. 

Quick question - - As it turns out in the case of the Datron 4911, simply put, there are five internal battery backups at 18V and 2 Amps each.  The assortment I ended up purchasing (as the originals are no longer made - long story) doesn't quite work out.  What I'm thinking of doing, but I don't have a schematic to prove whether it is feasible or not...

I wonder if it would be feasible to parallel two separate battery charge circuits together.  Because I couldn't get the right batteries, the way it is supposed to be set up is 4 battery sets at 18V / 2 AH.  What I have are two battery sets at 18V / 6 AH.  I determined that lows are all tied together, so that's not a problem.  But the charging lines on the

• side of each set are separate.  I'm reverse engineering this, so I'm looking for comments (unless someone has a 4910/4911 service manual).  There is apparently a separate charge and output circuit for each 18V battery/supply (a UPS supply for the four 10V references).  Is it reasonable to tie two of the 18V
• lines together, such that (for example) the UPS line for Cells 1 and 2 are connected together in parallel?  They are both 18VDC, and are supposed to go to two separate sets of batteries.  If I am substitting one high current battery bank for two lower current banks, would it work to have the two charge lines from two separate supply circuits tied together?
  
  I'm presuming not, and I don't want to destroy the DC standard.  Just an idea I'm tossing around.
  
  I've been waiting months to get my batteries.  Oh well.  I guess I'll have to get some more batteries and set them up external to the Datron 4911.
  
  The array is a bunch of 6V batteries.  I tried connecting two batteries per 10V reference cell (12V each), but it needs that 18V, and 12V doesn't quite cut it.  Problem I get with my severe ADD is my brain gets exhausted by this time of day.
  
  Thanks all for any thoughts.

Quote from: Hawaii596 on 03-14-2011 -- 16:31:44
I finally got my battery shipment in from China.  The product is fine and low cost.  As I learned, most all SLA (Sealed Lead Acid) batteries are made offshore - best I can determine, there are no OEM's for most SLA's in the US any more.  What they do is private label, etc.

If I had it to do over again, I think I'd prefer not to. 

Quick question - - As it turns out in the case of the Datron 4911, simply put, there are five internal battery backups at 18V and 2 Amps each.  The assortment I ended up purchasing (as the originals are no longer made - long story) doesn't quite work out.  What I'm thinking of doing, but I don't have a schematic to prove whether it is feasible or not...

I wonder if it would be feasible to parallel two separate battery charge circuits together.  Because I couldn't get the right batteries, the way it is supposed to be set up is 4 battery sets at 18V / 2 AH.  What I have are two battery sets at 18V / 6 AH.  I determined that lows are all tied together, so that's not a problem.  But the charging lines on the

• side of each set are separate.  I'm reverse engineering this, so I'm looking for comments (unless someone has a 4910/4911 service manual).  There is apparently a separate charge and output circuit for each 18V battery/supply (a UPS supply for the four 10V references).  Is it reasonable to tie two of the 18V
• lines together, such that (for example) the UPS line for Cells 1 and 2 are connected together in parallel?  They are both 18VDC, and are supposed to go to two separate sets of batteries.  If I am substitting one high current battery bank for two lower current banks, would it work to have the two charge lines from two separate supply circuits tied together?
  
  I'm presuming not, and I don't want to destroy the DC standard.  Just an idea I'm tossing around.
  
  I've been waiting months to get my batteries.  Oh well.  I guess I'll have to get some more batteries and set them up external to the Datron 4911.
  
  The array is a bunch of 6V batteries.  I tried connecting two batteries per 10V reference cell (12V each), but it needs that 18V, and 12V doesn't quite cut it.  Problem I get with my severe ADD is my brain gets exhausted by this time of day.
  
  Thanks all for any thoughts.
  

I apologize for resurrecting old threads, but this one came up in a search!  What did you end up doing with the battery situation?  Did you get your 4910 up and running?

I did get some batteries from China, and saved some in the price per unit. But in retrospect, for the little savings I got out of the efforts, I have decided to just buy domestically sold batteries. The 4910 ended up "out to pasture."  I currently have two active 732A's and two active 732B's. The 732B's I cal as secondary standards against a master 732A (that I do a "Hot Cal" via Fluke DVMP). I am at 0.08 ppm MU on my cal'd value (0.36 ppm over the course of the year) - and within 0.08 ppm on the pre-cal proficiency test.

Quote from: Hawaii596 on 05-05-2017 -- 09:04:38
I did get some batteries from China, and saved some in the price per unit. But in retrospect, for the little savings I got out of the efforts, I have decided to just buy domestically sold batteries. The 4910 ended up "out to pasture."  I currently have two active 732A's and two active 732B's. The 732B's I cal as secondary standards against a master 732A (that I do a "Hot Cal" via Fluke DVMP). I am at 0.08 ppm MU on my cal'd value (0.36 ppm over the course of the year) - and within 0.08 ppm on the pre-cal proficiency test.

I am glad to hear you were able to find some batteries for it and put it to some use, for however long you did.  The 4910 is my favorite DC reference and I hate to see them sit in corners unused .  I am actually in the process of designing my own LTZ1000 oven based reference.

Since you've retired it, any chance you would be willing to let it go?  I would love to put it in my home lab

I have two 4910 with two fully different batteries arrangements :
the first one has the classical MP2-6 batteries with foam forms to seat the batteries and a plate separation between the 2 levels of batteries coupled in 3X6V units. Easy to replace ;
the second one (german Army option 50) has 45 (!) US Enersys 0810-0004. They are grouped in plastic boxes with 6 X 12V, and 3 X 6V. The 6th 12V box is splitted into 2 X 6V to get 18V with the 4th and 5th 12V boxes. There is no separation in the battery box, no foam and all boxes are verticaly slipped (with force !), the 3 X 6V between the 12V rows, the 12V row under the connector being shorter. More interesting : every 12V box has a 10 KOhm thermistor (negative) glued on a single cell connected between +12V and a dedicated wire of the harness. I don't know if the purpose is voltage regulation of the individual charger with temperature) or temp sensing on the rear connector. Nothing on the schematics diagram, quite different from the real layout of the main PCB, and no time to investigate further : mounting of the 45 cells  and 5 resistors was a nightmare.
Beware of the charging voltage of the 5 individual 18V chargers : was to high with the Enersys specs !

Quote from: JFP on 12-02-2017 -- 05:51:16
I have two 4910 with two fully different batteries arrangements :
the first one has the classical MP2-6 batteries with foam forms to seat the batteries and a plate separation between the 2 levels of batteries coupled in 3X6V units. Easy to replace ;
the second one (german Army option 50) has 45 (!) US Enersys 0810-0004. They are grouped in plastic boxes with 6 X 12V, and 3 X 6V. The 6th 12V box is splitted into 2 X 6V to get 18V with the 4th and 5th 12V boxes. There is no separation in the battery box, no foam and all boxes are verticaly slipped (with force !), the 3 X 6V between the 12V rows, the 12V row under the connector being shorter. More interesting : every 12V box has a 10 KOhm thermistor (negative) glued on a single cell connected between +12V and a dedicated wire of the harness. I don't know if the purpose is voltage regulation of the individual charger with temperature) or temp sensing on the rear connector. Nothing on the schematics diagram, quite different from the real layout of the main PCB, and no time to investigate further : mounting of the 45 cells  and 5 resistors was a nightmare.
Beware of the charging voltage of the 5 individual 18V chargers : was to high with the Enersys specs !

Wow!  I've never seen a 4910 with the option 50.  Those batteries sound like a pain to replace.  I recently picked up a 4911 and the battery compartment wasn't in the greatest shape, so I am in the process of getting that rebuilt.  One of the yellow heat shrunk connectors needed to replace, and as you probably know, that is where the thermistor resides.  I believe the ones in ours 4 kOhm NTC.  Once I got the connector and thermistor replaced and all the batteries wired up, I was still getting a red 'battery supply' light, while each individual cell was getting the green 'battery light'.  I knew something had to be up with the charge circuitry.  I checked the charge voltages through the back 15pin connector and several of the banks were >22.5 V.  Closer to 24 V.

While I've got the unit torn apart, I'm changing a bunch of the tantalum and electrolytic caps so, hopefully, I shouldn't have to open it back up for another decade or so. 

I would like to ask, what did you end up doing to set the charge voltage?  I was thinking it might be the blue/gray pots on the PSU module

Typing error for the option : 90Y
I have the schematics and layout of motherboard DC400878 where the pots a grouped in 2 x5 separeted by some space : there are pots for the "Set charge" 'RX24 with X=1 to 5) and pots for the "Cm adjust" "RX01...). No instruction about the procedure to adjust the common mode rejection ((100K between AC going to the bridge rectifier, the wiper going to the guard and the shield between primary and secondary of main transformer).
With this board the charge adjustement pots are the second group from the rear , beginning with battery 5 then 1, 2, 3 and 4 the last pot towards the front of the instrument.
For the motherboard 400971 the pots are in a group of 10 with different numbers : RX07 for charge voltage adjust and RX04 for CM adjust. The first pot at the rear is for battery 1, then 2,3,4,5 and then reverse for the CM 5,4...1 the last pot to the front of the instrument.
Setting the charge voltage limit is tricky and involves battery out, 10Kohm betwen X and Y pins of battery connector for battery 1 and voltage across RX10, 3,4535 +-1mV for isue 400971 A and 3,6937... for isue I or later. I have no schematics, no layout and my procedure was more rough : voltage across battery 1 and pot adjust to the limit given by the battery data

You said: "The 4910 is my favorite DC reference..." I wonder if you have some thoughts about that. I have basically mothballed the 4910, and I am using a 732A as the main house standard (Fluke "Hot Cal" every year), and 2 732B's and a second 732A as comparison standards (and the technicians use the 732B's as portable references for calibrating HP 3458A's etc. with them).

Seemed as though I had plotted the 4910 for a while, and the drift numbers weren't as good as the Fluke units. Maybe each zener has its own characteristic drift pattern.

Quote from: Hawaii596 on 12-13-2017 -- 17:33:56
You said: "The 4910 is my favorite DC reference..." I wonder if you have some thoughts about that. I have basically mothballed the 4910, and I am using a 732A as the main house standard (Fluke "Hot Cal" every year), and 2 732B's and a second 732A as comparison standards (and the technicians use the 732B's as portable references for calibrating HP 3458A's etc. with them).

Seemed as though I had plotted the 4910 for a while, and the drift numbers weren't as good as the Fluke units. Maybe each zener has its own characteristic drift pattern.

My reasoning behind the 4910 being my favorite DC reference is due to the fact it implements a PWM DAC for amplification rather than normal op amplification in the 732A and B.  This type of implementation is more intriguing to me.

Brand new, I would think each type of zener would exhibit similar drift patterns.  Ofcourse some batches/specimens will yield better results than others.  But with as old as these units are, and with the entire history of each zener being unknown, it's hard to say 'X is better than Y' because it not only depends on the zener itself, but how well the unit was maintained and cared for.  One could have had a reference that started out rock solid stable, but then maybe it was sold and stored unpowered for a significant amount of time. 

While 732As are generally the most finicky of the 3, they do seem to be the most stable.  That could be for a multitude of reasons, though.  They were produced before the 4910 and therefore have a longer history to allow the zener to settle. 

In my home lab I've got a 732A and 732B.  The 732A I picked up a few months back, off the bay, with unknown history and it is still a little noisy ~2 µVp-p and drifting up still.  It has been slowing down, but still drifting significantly compared to the references at work. 

Here at work we've got a 4910, 4911, and 732B.  The 732B we've had for 5+ years and have established a decent history on.  The 4910 I've been monitoring for about a year now and it has been rock stable.  The 4911 was recently picked up and is still in the process of being refurbished.  I haven't had the 4910 or 4911 for long enough to be able to speak to their drift characteristics.  Here in another year or 2 I will be more certain with everything.