Re: [volt-nuts] DAS-46

[Jerry Hancock]

I just sent Charles a note off-list.

This morning I received replacement FETs for the MFE3002 using ECG220s.  These 
came with a little spring around the leads keeping the leads shorted as they 
are high impedance.  I replaced the suspect FET and the unit is working!  Go 
figure, as I suspected I was moving the problem around and I guess the first 
FET I put in was bad.

Anyway, I appreciate all the support and help provided by such an august group 
of members and will send along a plot of the device once it warms over a few 
days.

Jerry
___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

Re: [volt-nuts] DAS-46 CSA cost

[Orin Eman]

Actually, not too bad considering they probably have to dig out the
components and build/test one by hand.  If it weren't so much, it would be
interesting to see whether they have redesigned it to use more modern
components.



On Mon, Jan 30, 2017 at 11:02 AM, Jerry Hancock  wrote:

> I received a note from Prime Technology today and the CSA will cost $425
> with a lead time of 10 weeks. Ha!  I can purchase 3 or 4 of them for that
> price.
>
> Anyway, I don’t blame them.
> ___
> volt-nuts mailing list -- volt-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/volt-nuts
> and follow the instructions there.
>
___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

Re: [volt-nuts] DAS-46

[Charles Steinmetz]

Jerry wrote:


One of the items I am having trouble modeling is a CLD, current limiting diode, 
MCL1302.


CLDs are almost always operated in saturation, so a behavioral current 
source of the appropriate value should model it just fine for your needs 
(spec for the MCL1302 is 2mA +/- 0.6mA -- as you can see, hardly 
precision devices).


CLDs are nothing more than JFETs operating at Idss, so you can also 
model them with a JFET with its gate tied to its source. Of course, you 
have to try different FET models until you find one with the appropriate 
Idss, or play with source resistors to lower the saturation current of a 
FET with somewhat higher Idss.


But if you insist, Central Semi has models of its CLDs on its web site:



Central makes a direct substitute (the CMCL1302), but does not seem to 
have a Spice model for that particular one.  The CCL2000 is another 2mA 
CLD, with similar specs (but a somewhat tighter current variance spec). 
 You can find its P-Spice model here:




I'm sure that model would be more than close enough for your purposes.

Best regards

Charles


___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

Re: [volt-nuts] DAS-46

[Jerry Hancock]

Well, you aren’t going to believe this, but I’ve checked the circuit about 10 
times now and it looks pretty good.  I then used Charles advice and imported 
the CCL2000 CLD.  I ran the model and it swung to the negative rail to almost 
the exact voltage I am getting with the real circuit of -13.979V.I know I 
still have errors in the model but I find this to be pretty funny.  


___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

Re: [volt-nuts] DAS-46

[Jerry Hancock]

Charles, thank you!  Very helpful, as everyone has been.


> On Jan 29, 2017, at 6:38 PM, Charles Steinmetz  wrote:
> 
> Jerry wrote:
> 
>> One of the items I am having trouble modeling is a CLD, current limiting 
>> diode, MCL1302.
> 
> CLDs are almost always operated in saturation, so a behavioral current source 
> of the appropriate value should model it just fine for your needs (spec for 
> the MCL1302 is 2mA +/- 0.6mA -- as you can see, hardly precision devices).
> 
> CLDs are nothing more than JFETs operating at Idss, so you can also model 
> them with a JFET with its gate tied to its source. Of course, you have to try 
> different FET models until you find one with the appropriate Idss, or play 
> with source resistors to lower the saturation current of a FET with somewhat 
> higher Idss.
> 
> But if you insist, Central Semi has models of its CLDs on its web site:
> 
> 
> 
> Central makes a direct substitute (the CMCL1302), but does not seem to have a 
> Spice model for that particular one.  The CCL2000 is another 2mA CLD, with 
> similar specs (but a somewhat tighter current variance spec).  You can find 
> its P-Spice model here:
> 
> 
> 
> I'm sure that model would be more than close enough for your purposes.
> 
> Best regards
> 
> Charles
> 
> 
> ___
> volt-nuts mailing list -- volt-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
> and follow the instructions there.

___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

Re: [volt-nuts] DAS-46

[Jerry Hancock]

Yes, and I assume you would need an FET that acted like the CLD.  I believe the 
MCL1302 is a 2mA device.  I doubt I’[ll get the model working, which would be 
ideal.

> On Jan 29, 2017, at 3:20 PM, Bruce Griffiths  
> wrote:
> 
> A CLD is merely a JFET with the gate connected to the source.Sometimes a 
> source resistor may be included with the gate connected to the opposite end 
> of the resistor to the end connected to the source.
> Bruce 
> 
>On Monday, 30 January 2017 12:12 PM, Jerry Hancock  
> wrote:
> 
> 
> One of the items I am having trouble modeling is a CLD, current limiting 
> diode, MCL1302.  The CLDs are used between he base and collectors on the 
> output stage 2n2219 / 2n2905.  I swapped them on the board with no impact on 
> the problem.  I can’t find CLDs in LTSPICE.
> 
> The other diode I can’t read and it is used cathode to cathode in a weird 
> arrangement.  The diode is roughly 1/2 the length (in this case height) of a 
> 1n4148.  It measures as a diode with .51V forward.  I have a hard time 
> telling if it actually has a cathode band.  I haven’t pulled it out but I 
> guess I’ll have to and put it under the microscope to see if there is a 
> number on it.  I have a lot of very low forward voltage switching diodes and 
> Schottky type so I probably have it on-hand if I can get a number off of it.
> 
> I spent some more time on the circuit today.
> 
> By the way, even though the Parent company, was General Resistance, now Prime 
> Technology, said they would sell me a CSA, they said they won’t share the 
> schematic.
> 
> Thanks again.
> ___
> volt-nuts mailing list -- volt-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
> and follow the instructions there.
> 
> 
> 
> ___
> volt-nuts mailing list -- volt-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
> and follow the instructions there.

___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

Re: [volt-nuts] DAS-46

[Bruce Griffiths]

A CLD is merely a JFET with the gate connected to the source.Sometimes a source 
resistor may be included with the gate connected to the opposite end of the 
resistor to the end connected to the source.
Bruce 

On Monday, 30 January 2017 12:12 PM, Jerry Hancock  wrote:
 

 One of the items I am having trouble modeling is a CLD, current limiting 
diode, MCL1302.  The CLDs are used between he base and collectors on the output 
stage 2n2219 / 2n2905.  I swapped them on the board with no impact on the 
problem.  I can’t find CLDs in LTSPICE.

The other diode I can’t read and it is used cathode to cathode in a weird 
arrangement.  The diode is roughly 1/2 the length (in this case height) of a 
1n4148.  It measures as a diode with .51V forward.  I have a hard time telling 
if it actually has a cathode band.  I haven’t pulled it out but I guess I’ll 
have to and put it under the microscope to see if there is a number on it.  I 
have a lot of very low forward voltage switching diodes and Schottky type so I 
probably have it on-hand if I can get a number off of it.

I spent some more time on the circuit today.

By the way, even though the Parent company, was General Resistance, now Prime 
Technology, said they would sell me a CSA, they said they won’t share the 
schematic.

Thanks again.
___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.


   
___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.

Re: [volt-nuts] DAS-46

[Charles Steinmetz]

Jerry wrote:


I traced the issue to a Chopper Stabilized Amp on the back of the power supply 
board.

Unfortunately, the CSA has no schematic supplied.  If it fails, it is to be 
replaced.
I’ve been trying to reverse engineer it into LTSpice without luck.
There are two FETs, MFE3002 that cross to ECG220


Be careful of ECG and NTE parts -- sometimes they are so dissimilar to 
the crossed part that one wonders how on earth someone thought they were 
useful substitutes.  I have data on the MFE3002, but not on the ECG220, 
so I can't speak re: this particular pairing.  I can tell you that 
Mouser doesn't have anything even remotely similar (but I suspect that 
some of the parts they do have can be made to work).


Alternatively, lots of great chopper-stabilized amplifiers are available 
as ICs these days, so you might be able to replace the whole module with 
one IC and a few other parts.


Do try to repair what you have -- if you post a schematic when you are 
done tracing the circuit, we can help you get it working (if it isn't 
obvious by then what is wrong).


I'm attaching the datasheet for the MFE3002 in case you don't have it.

Best regards,

Charles


___
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.