22-05-2011, 09:09 PM
(This post was last modified: 01-11-2011, 11:05 AM by Mark Hennessy.)
Despite having various bench-type PSUs knocking around, these seem to be the one item of test gear that I never seem to have enough of - or rather, one that comes to hand & that gives me the voltages and currents that I require, preferably in one box.
So during a tidy-up a few weeks ago, when I decided that I would have a closer look at a redundant PSU that I kept tripping over, I realised that this could provide a basis for a solution. In essence, the transformer and case were 're-usable'; new 'innards' would be required. The transformer is a C-core type with several secondary windings and by a suitable inter-connection of some of these plus a modification to another & subsequent test - the project was launched!
That's how it all started; it's now completed; hence this write-up.
Whenever I do design & build something for myself, I always make detailed notes & drawings of what it is I've created: useful for future maintenance and reference. So, rather than give a blow-by-blow account of what happened and what I ended up with, etc., here I've simply 'pasted in' these notes and pictures - which hopefully tells enough, for those who are interested.
General design comments.
The overall design approach was based on the following concepts.
1. It began with “Here’s a useful - looking transformer & case. Without spending any money (or as little as possible) but by using parts that I have (electrical & mechanical; salvage & second-hand), what can I put together to make something useful?”. Various electrical & mechanical designs were investigated: the final design was arrived at by a process of ‘read & research, analyse, build, test, modify & evolve’, with the usual engineering trade-offs and compromises. This part of the whole project involved the greatest amount of time.
2. Reliability and ease of maintenance were given a high priority - even if this meant compromising some performance characteristics. To this end, on the mechanical front, the unit has been built as three separate assemblies (plus case): front panel, rear panel with heat-sinks & transformer ass’y. This arrangement enables all of the electronics to be removed entirely from the case, re-connected and switched on. On the electrical front, over-rated components were used (where possible) - e.g. the large heat-sinks on the back panel and the use of robust power Darlington transistors for the series-pass elements of the variable-voltage regulators. The ‘skeleton’ style of construction - based on the transformer providing the ‘frame’ - was deliberately chosen to facilitate ease of maintenance (as previously mentioned).
Transformer & rectification.
1. One sec’y. wndg. was originally 21v - 0 - 21v, rated at 1.5 amp. This was modified to produce two separate 21v. wndgs. Two bridge rectifiers are used.
2. Two 6v - 0 - 6v wndgs. are connected in series, (phase-additive); two bridge rectifiers are used.
3. A sec'y. wndg. of 115v @ 1.5 amp. is not used.
4. The pri. wndg. has taps at 200, 220 and 240 v. The incoming mains supply is connected to the 220v. tap.
Case.
Replacement rear & front panels were produced, drilled and painted. The case was re-painted. All paint was readily available as ‘left-overs’ from earlier projects.
Variable O/Ps voltage regulators
1. Although labelled as “+ve O/P” & “-ve O/P”, these two O/Ps are floating above chassis / earth and are completely independent of each other. The names are purely for ease of reference.
2. Each regulator cct. has an automatic electronic over-current trip that switches off the O/P voltage at currents exceeding approx. 2.6 amps. This trip is primarily indented as a protection device for the PSU, not, primarily, as protection for any external load. The cut-out is of the ‘fold-back’ type, not the ‘constant-current’ type.
3. The over-current protection does not change when the 300 mA range is selected on the front panel ammeter. Therefore caution should be exercise when this low current meter range is selected. However, it was found that dramatic overloads on the O/P did not cause permanent damage to the meter when the 300 mA range was selected.. Nevertheless, this ‘feature’ should not be relied upon when low currents are being supplied.
4. Each of these two outputs should be regarded at having a max. Vout of 20 volts. Although the O/P will go to 22v, the regulation and noise performance is degraded above 20v. At 20v O/P (or less) the peak-peak noise components are at least -50dB relative to the d.c. voltage O/P at currents up to 2 amps. The minimum Vout is 2.5 v. The O/P voltage is adjusted by a 10-turn precision pot.; the resultant O/P voltage is displayed on the switched front-panel mounted voltmeter.
Fixed O/P voltage regulators.
The ccts. used are simply the classic configurations using TO-3 fixed-voltage regulator int. ccts. Upon test, each O/P was found to exceed its rated current O/P for several minutes with an acceptably low noise level.
----------------------
And that'll do for now. I could go into more detail - but I'll wait for any responses; thanks for wading through that lot above!
Al. / Sunday, May 22nd. //
P.S. If it's not obvious, pic. 4 is the inside view of the front panel.
So during a tidy-up a few weeks ago, when I decided that I would have a closer look at a redundant PSU that I kept tripping over, I realised that this could provide a basis for a solution. In essence, the transformer and case were 're-usable'; new 'innards' would be required. The transformer is a C-core type with several secondary windings and by a suitable inter-connection of some of these plus a modification to another & subsequent test - the project was launched!
That's how it all started; it's now completed; hence this write-up.
Whenever I do design & build something for myself, I always make detailed notes & drawings of what it is I've created: useful for future maintenance and reference. So, rather than give a blow-by-blow account of what happened and what I ended up with, etc., here I've simply 'pasted in' these notes and pictures - which hopefully tells enough, for those who are interested.
Multi O/P low-voltage bench PSU
General design comments.
The overall design approach was based on the following concepts.
1. It began with “Here’s a useful - looking transformer & case. Without spending any money (or as little as possible) but by using parts that I have (electrical & mechanical; salvage & second-hand), what can I put together to make something useful?”. Various electrical & mechanical designs were investigated: the final design was arrived at by a process of ‘read & research, analyse, build, test, modify & evolve’, with the usual engineering trade-offs and compromises. This part of the whole project involved the greatest amount of time.
2. Reliability and ease of maintenance were given a high priority - even if this meant compromising some performance characteristics. To this end, on the mechanical front, the unit has been built as three separate assemblies (plus case): front panel, rear panel with heat-sinks & transformer ass’y. This arrangement enables all of the electronics to be removed entirely from the case, re-connected and switched on. On the electrical front, over-rated components were used (where possible) - e.g. the large heat-sinks on the back panel and the use of robust power Darlington transistors for the series-pass elements of the variable-voltage regulators. The ‘skeleton’ style of construction - based on the transformer providing the ‘frame’ - was deliberately chosen to facilitate ease of maintenance (as previously mentioned).
Transformer & rectification.
1. One sec’y. wndg. was originally 21v - 0 - 21v, rated at 1.5 amp. This was modified to produce two separate 21v. wndgs. Two bridge rectifiers are used.
2. Two 6v - 0 - 6v wndgs. are connected in series, (phase-additive); two bridge rectifiers are used.
3. A sec'y. wndg. of 115v @ 1.5 amp. is not used.
4. The pri. wndg. has taps at 200, 220 and 240 v. The incoming mains supply is connected to the 220v. tap.
Case.
Replacement rear & front panels were produced, drilled and painted. The case was re-painted. All paint was readily available as ‘left-overs’ from earlier projects.
Variable O/Ps voltage regulators
1. Although labelled as “+ve O/P” & “-ve O/P”, these two O/Ps are floating above chassis / earth and are completely independent of each other. The names are purely for ease of reference.
2. Each regulator cct. has an automatic electronic over-current trip that switches off the O/P voltage at currents exceeding approx. 2.6 amps. This trip is primarily indented as a protection device for the PSU, not, primarily, as protection for any external load. The cut-out is of the ‘fold-back’ type, not the ‘constant-current’ type.
3. The over-current protection does not change when the 300 mA range is selected on the front panel ammeter. Therefore caution should be exercise when this low current meter range is selected. However, it was found that dramatic overloads on the O/P did not cause permanent damage to the meter when the 300 mA range was selected.. Nevertheless, this ‘feature’ should not be relied upon when low currents are being supplied.
4. Each of these two outputs should be regarded at having a max. Vout of 20 volts. Although the O/P will go to 22v, the regulation and noise performance is degraded above 20v. At 20v O/P (or less) the peak-peak noise components are at least -50dB relative to the d.c. voltage O/P at currents up to 2 amps. The minimum Vout is 2.5 v. The O/P voltage is adjusted by a 10-turn precision pot.; the resultant O/P voltage is displayed on the switched front-panel mounted voltmeter.
Fixed O/P voltage regulators.
The ccts. used are simply the classic configurations using TO-3 fixed-voltage regulator int. ccts. Upon test, each O/P was found to exceed its rated current O/P for several minutes with an acceptably low noise level.
----------------------
And that'll do for now. I could go into more detail - but I'll wait for any responses; thanks for wading through that lot above!
Al. / Sunday, May 22nd. //
P.S. If it's not obvious, pic. 4 is the inside view of the front panel.
And that consideration includes future repairs and reworks. To that end, I have included the schematics below. I'm sure that there will be some improvements that could be made therein & I welcome any constructive comments from anyone. After all, I'm not an Engineer, merely a humble Technician; I repair stuff, not design it.
- easily retrofitted.
