B&K Precision 1671A - very coarse voltage adjustment!

[Chalky]

After seeing the 1670A on Dave's Amazon list, I thought I'd go for the 5A model - the 1671A. 

Unit seems to work OK, but the voltage adjustment is VERY coarse, jumps about a bit from the slightest touch.  Quite hard to even keep it within half a volt of the desired setting.  Voltage goes up to 30, so using a single three-quarter turn carbon pot was probably not the best idea B&K ever had.  Pot is marked CTS 50KD and 1209 - am guessing the 50KD is the important bit, it measures 45k end-to-end.  Will try to replace it it something better, would love to find a multiturn pot with same form factor.

Cheers, hope this can be of some use to someone, will post any further findings that seem relevant.

Refs:
http://www.bkprecision.com/products/power-supplies/1671A-triple-output-30v-5a-digital-display-dc-power-supply.html
http://www.ctscorp.com/components/Datasheets/250.pdf

[LaurenceW]

It's probably got a linear pot, so that each 10 degrees of rotation is worth roughly the same voltage increment (roughly 1Volt). The problem is, you are more bothered about setting 3.3V +/- 0.1 volts, than you are about setting (say) 25V +/- 0.1V. Ideally, you'd like better resolution of the voltage setting at lower volts, in exchange for poorer resolution at higher settings. One answer to this might be to use a logarithmic, in place of the current linear, 50K pot. However, you are apt to find that the log-law pot "overdoes it" and that maybe half or the pot's rotation covers 0 - 4V, and the other half rotation covers the remaining 26V!

Try this. Strap a resistor (start with 10K, but you will need to experiment 2K7 > 47K) between the centre/wiper pin of the pot, and the cold/ground side. This will distort the linear behaviour of the pot to some degree, giving more degrees of rotation to the lower voltage end of the pot's adjustment.

If you don't mind drilling an extra hole, try adding a small toggle switch alongside the voltage pot for "high/Low" settings. Arrange for this to switch in an additional 100K in series with the high side of the pot, thus giving you a 0>10V range on the low setting.

While you have got the drill out, you might like to think about adding an additional output on/off switch. This allows you to check the voltage rating, before accidentally applying it to your unsuspecting circuit...

Ten turn pots sound great, until you find the supply set on 30V when you wanted 3.3V. That's a lot of twiddling...

[SeanB]

You do get multiturn pots that are 3 turns, so almost are the same size as a regular one.

[MikeK]

Interestingly, 3-turn pots seems to be more expensive than 10-turn.  I would much rather have a 3- or 5-turn pot than a 10-turn.  I would have thought 10-turns to be more expensive. ??

[poida_pie]

This BK PS seems a lot like my Mason NP-9615
http://www.altronics.com.au/index.asp?area=item&id=M8205
which I bought 6 months ago.

The first thing I did when I got it home was take it apart.
(Dave has taught us well...)
The next thing was note the pots and get some 10 turn versions of the two and fit them a couple of days later.

10 turn pots make this unit a lot nicer. They also are quick to get from one end to the other if you just
push a finger past on one side, turning the pot a few turns each swipe.

[SeanB]

3 turn units are a lot slower selling.

[Chalky]

Thanks guys, all great suggestions.  I have implemented a solution, but I might do the output switch thing - I really like that idea, but will see how I go with the banana leads.  Anyway, here's what I did:

I was hoping to find the pot was being used as a straight voltage divider, as the only 10-turn I could get quickly was 100k (versus original 50k).  Unfortunately, only one 'end' of the pot is used, i.e. and was paralleled with a single fixed value 470k resistor, give a range of about 0 to 41K . So, replaced the 50k pot with 100k 10-turn pot, did some quick maths and trial and error, added another ~80k resistor in parallel, this gives me a maximum output voltage of just on 30.1V.  Perfect!

Interestingly, the original variable and fixed resistors were of quite low tolerance, and the original output voltage went up beyond 32V.  No sign of a trimpot to control max voltage, hence my trial and error with fixed value, above.

Cheers.

[Chalky]

This BK PS seems a lot like my Mason NP-9615
http://www.altronics.com.au/index.asp?area=item&id=M8205
which I bought 6 months ago.

The first thing I did when I got it home was take it apart.
(Dave has taught us well...)
The next thing was note the pots and get some 10 turn versions of the two and fit them a couple of days later.

10 turn pots make this unit a lot nicer. They also are quick to get from one end to the other if you just
push a finger past on one side, turning the pot a few turns each swipe.

Hey Poida Pie, does look the same, can you tell me (send link?) for the pot you used?  50k?

[poida_pie]

Chalky

I used the pots from rs components, the 50 K item is

http://australia.rs-online.com/web/p/potentiometers/1070808/?searchTerm=107-0808

(the RS component stock no. is 107-0808 for the 50K, 107-0779 for the 5K pot)

I found these 10 turn pots were just a little too deep to fit in the original holes, so I made two new holes nearby
on the front panel. I intend to plug the old unused holes one day but we all know that's never going to happen...

$18 per pot, free delivery for me in AUS.

[Chalky]

Yeah my pot was too wide, I had to shave the side of one of the plastic mount posts for the small PCB!  However, end result is really great, I can get 0.1 volt control without any effort whatsoever.