FMC DIO 5ch TTL a issueshttps://ohwr.org/project/fmc-dio-5chttla/issues2019-02-12T09:55:21Zhttps://ohwr.org/project/fmc-dio-5chttla/issues/18Too big RDSon in mosfets2019-02-12T09:55:21ZTomasz WlostowskiToo big RDSon in mosfetsReplace the termination MOSFETs (BST82) with models with low RDSon @ Vgs
== 2.5V, for example BSH103.Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/17Max buffer current output too low2019-02-12T09:55:20ZTomasz WlostowskiMax buffer current output too lowThe output buffers can't properly drive a 50-ohm load. The current
output should be increased, possibly by putting two '125 drivers in
parallel.https://ohwr.org/project/fmc-dio-5chttla/issues/16Missing copper fill next to the front panel2019-02-12T09:55:19ZTomasz WlostowskiMissing copper fill next to the front panelThe copper fill (polygon) close to the front panel mounting hole is
missing.https://ohwr.org/project/fmc-dio-5chttla/issues/151-wire thermometer chip Vih 0.1V too high.2019-02-12T09:55:18ZErik van der Bij1-wire thermometer chip Vih 0.1V too high.Regarding the 1-Wire serial number / thermometer chip, the Maxim
DS18B20U+:
The data sheet says that when operated from a 3.3V supply (2.5V is not
possible), the minimum input high voltage level (Vih) is 2.2V.
On the other hand the Spartan 6 (that is used on the SPEC) data sheet
says that its minimum output high voltage level (Voh) on a 2.5V powered
bank is 2.1V. So it's close but not ideal.
- Verify above statement
- Decide if is acceptable or not.Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/14V1 - Problem with solder mask2019-02-12T09:55:17ZErik van der BijV1 - Problem with solder maskThere are 4 vias on the thermal pad of the IC11 on the FMC board (see in
attached figure) which are uncovered with anti-soldering mask on the
"etm" layer and covered on the "ebm" layer.
This create an effect called "ladle" which transports chemical product
between different bath.
- Recommend to uncover these vias on the "ebm" layer with 0.35mm
- This recommendation is implemented by the PCB fabricator on the
first production batch (4790415)
### Files
* [Mask.jpg](/uploads/57b90df7e699348a222db1440926d494/Mask.jpg)Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/13Wrong type of IC2 and IC5 in EDMS release2019-02-12T09:55:16ZTomasz WlostowskiWrong type of IC2 and IC5 in EDMS releaseIn the EDMS release, IC2 and IC5 are of SN74LS125D type, but should be
of SN74LVT125D. This causes too low hi-level voltage and too slow edges
in output signals.Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/12V1-1 - wrong manufacturer of DAC55782019-02-12T09:55:16ZTomasz WlostowskiV1-1 - wrong manufacturer of DAC5578The manufacturer field of DAC5578 should be "Texas Instruments", but is
"Analog Devices".Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/11V1-1 - BOM Reference FMC connector wrong (non-RoHS)2019-02-12T09:55:15ZErik van der BijV1-1 - BOM Reference FMC connector wrong (non-RoHS)The reference of the FMC connector is wrong as it is the Tin-Lead
version instead of the Lead-Free version. See Samtec SEARAY VITA 57
connectors.
Reference used : MC-LPC-10L - ASP-127797-01 (Tin-Lead)
Correct reference: MC-LPC-10 - ASP-134604-01 (Lead-Free)
Actions
- Correct schematics: Connector J1 to become ASP-134604-01
- Make type visible in schematics (currently not shown).
<!-- end list -->
- Create V1-2 (no PCB layout change needed) or V2-0 if additional PCB
changes are required.
- Update production files
This [Issue also appeared on the
TDC](https://www.ohwr.org/work_packages/457) .Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/10V1-1 - Pull up resistors not strong enough to disable outputs at startup2019-02-12T09:55:14ZErik van der BijV1-1 - Pull up resistors not strong enough to disable outputs at startupThe 10 kOhm pull up resistors (R1, R11a/b, R12a/b) are not strong enough
to disable outputs at power up. Therefore the design is out of spec.
Issue found by Seven Solutions.
Specification of outputs:
- Output levels: LVTTL, capable of driving +3.3 V over a 50-Ohm load.
**At power-up the outputs should be in Hi-Z state**
- Withstands a continuous short-circuit on all the outputs at the same
time
Action:
- Replace R1, R11a/b and R12a/b by 4k7
### Files
* [Mail.txt](/uploads/90f602832f8886606a87d05e8b0f302b/Mail.txt)
* [LVTH_mail.txt](/uploads/9908d39f003699e47866988089ebfc54/LVTH_mail.txt)
* [family.lvt.specification.pdf](/uploads/3d4fe5190ccc36e96f511797d24f6746/family.lvt.specification.pdf)
* [pull-up.txt](/uploads/fdd5ac58194d4f73ef9d4d0ce75a1901/pull-up.txt)Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/9V1-1 - Continuous short on output specification not valid?2019-02-12T09:55:13ZErik van der BijV1-1 - Continuous short on output specification not valid?The short specification reads: *"Withstands a continuous short-circuit
on all the outputs at the same time"*
Measurements made:
Measured the current supplied by single driver when its output is set to
a high state and it is connected to ground through a 30ohm resistor. The
current that flows through the resistor is about 70mA.
This is more than the “recommend operating conditions” value (32mA) and
more than the “absolute maximum ratings” value (64mA) of the 74LVT125.
Curiously, this “absolute maximum ratings” value is not specified in the
Texas Instruments' datasheet, however, it can be found in the
corresponding datasheet of other manufacturers of the same IC (NXP and
Fairchild).
- Remove this line in the specification?Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/8V1-1 - PCB layer stack unclear2019-02-12T09:55:12ZTomasz WlostowskiV1-1 - PCB layer stack unclearLayer 2 should be marked as POWER plane instead of SIGNAL LAYER.Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/7V1-1 - Vih of 5V will blow fuse2019-02-12T09:55:11ZErik van der BijV1-1 - Vih of 5V will blow fuseThe inputs are supposed to work with a Vih up to 5V (Specification:
Input levels: any logic standard from Vih = 1 V to Vih = 5 V).
However, the inputs are just protected by a fuse in series with a diode
connected to 3V3.
Therefore with any Vih higher than 4.0V the fuse may blow and the input
signal will not be correctly handled.
- Redo input circuit protection. Note that the same connector can also
be used as output.Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/6V1-1 - Add 4 screws in BOM2019-02-12T09:55:11ZErik van der BijV1-1 - Add 4 screws in BOMIt would be good if the FMC mezzanine would have the four screws to
assemble it to the carrier would be included in the package.
- Add to BOM: 4 x BN 5687 ISO 14583 - Hexalobular Pan Head Screws -
Fully Threaded - Stainless Steel A2Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/5V2 - clamping current of ADCMP6042019-02-12T09:55:10ZTomasz WlostowskiV2 - clamping current of ADCMP604When the card is unpowered and positive voltage is applied to the input,
the clamping is done by the protection circuits of ADCMP604 comparators
instead of BAR66 diodes. This in turn means too high clamping current
which may damage the comparator.
To Do: add 100 ohm series resistors on the positive comparator inputs.Tomasz WlostowskiTomasz Wlostowskihttps://ohwr.org/project/fmc-dio-5chttla/issues/4V2-0 - Padstack of 5 polyfuses is wrong2019-02-12T09:55:09ZErik van der BijV2-0 - Padstack of 5 polyfuses is wrongThe padstack of the TYCO ELECTRONICS miniSMDC014F Polyfuses F1, F2A,
F2B, F3A and F3B is wrong.
See attached file.
- Verify
- Request design office to correct the padstack of the symbol in the
library.
- Signal in EDMS a warning about the assembly of these polyfuses.
- Put Issue on hold until more issues arrive before creating a V3-0.
Remark: same Polyfuse used in [TDC
card](https://www.ohwr.org/project/fmc-tdc/wiki)
### Files
* [Polyfuse.pdf](/uploads/323cb9e9d1678e556cf444c8586b3f59/Polyfuse.pdf)
* [Tyco_IPC.pdf](/uploads/f75f79cc827da77427f1d2960792d228/Tyco_IPC.pdf)Tomasz WlostowskiTomasz Wlostowski